repo
stringclasses
10 values
pr_number
int64
7
155k
title
stringlengths
4
137
body
stringlengths
0
68.4k
buggy_commit
stringlengths
40
40
fix_commit
stringlengths
40
40
buggy_distance
int64
1
30
confidence
stringclasses
3 values
files
listlengths
1
5
vercel/next.js
88,794
[ci] Ensure Turbo Remote Cache can be written to
Follow-up to https://github.com/vercel/next.js/pull/88746 The token in `TURBO_TOKEN` has no write permissions. Unclear why Turbo logged "Remote caching enabled" if all the roles in Vercel either have read/write or no permissions. I added a new token under the `vtest314-next-e2e-tests` scope which is more relevant...
4379e65e5fdb0dc5e218be2b4711c5ff85b5cb3e
2ed21f26d3ccd9b3c6b829c4af01ec6f5e93e68f
12
medium
[ { "filename": ".github/workflows/build_reusable.yml", "patch": "@@ -95,9 +95,9 @@ env:\n # disable backtrace for test snapshots\n RUST_BACKTRACE: 0\n \n- TURBO_TEAM: 'vercel'\n+ TURBO_TEAM: 'vtest314-next-e2e-tests'\n TURBO_CACHE: 'remote:rw'\n- TURBO_TOKEN: ${{ secrets.TURBO_TOKEN }}\n+ TURBO_TOK...
vuejs/vue
9,154
fix(types): broaden VNodeChildren, fix #8498
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [x] Bugf...
448ba65d2b139b29f1e6891add9925ac22ffe10b
d780dd2e2adcf71f40c086055a659a9a2b4a8282
18
medium
[ { "filename": "types/test/options-test.ts", "patch": "@@ -281,6 +281,12 @@ Vue.component('provide-function', {\n })\n })\n \n+Vue.component('component-with-slot', {\n+ render (h): VNode {\n+ return h('div', this.$slots.default)\n+ }\n+})\n+\n Vue.component('component-with-scoped-slot', {\n render (...
ollama/ollama
9,483
Update README.md
removed solar and added granite 3.2 by ibm. redhat looking for using this on there linux os so having this one make sense and removed solar as it is now irrelevant not been updated and it is too out of date for the quality.
fefbf8f74bdedeb4a6066f0b958a52194de91240
8fe6f69f2828d66df7fafa02cf6ee5cb030ef87d
4
medium
[ { "filename": "README.md", "patch": "@@ -75,7 +75,7 @@ Here are some example models that can be downloaded:\n | Code Llama | 7B | 3.8GB | `ollama run codellama` |\n | Llama 2 Uncensored | 7B | 3.8GB | `ollama run llama2-uncensored` |\n | LLaVA | 7B | ...
ggml-org/llama.cpp
19,761
grammar : Fix grammar root symbol check
# The problem Constructing a GBNF grammar allows the programmer to select a `grammar_root`- the symbol to start the grammar from. The `llama_grammar_init_impl` function incldued a check to see whether the grammar contains a rule for a symbol named literally "root", instead of checking for a symbol with the named ...
null
0a10c34dc146b19cc256a66b36321d753899a684
null
low
[ { "filename": "src/llama-grammar.cpp", "patch": "@@ -1160,13 +1160,13 @@ struct llama_grammar * llama_grammar_init_impl(\n // if there is a grammar, parse it\n // rules will be empty (default) if there are parse errors\n if (!parser.parse(grammar_str) || parser.rules.empty()) {\n- fprintf...
vuejs/vue
7,647
fix(sfc): avoid to deindent when pad option is specified
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [x] Bugf...
76fd45c9fd611fecfa79997706a5d218de206b68
9d2f9a034f9c40d5ba6d8b1e131b1bfb675dc1cf
1
medium
[ { "filename": "src/sfc/parser.js", "patch": "@@ -83,11 +83,15 @@ export function parseComponent (\n function end (tag: string, start: number) {\n if (depth === 1 && currentBlock) {\n currentBlock.end = start\n- let text = deindent(content.slice(currentBlock.start, currentBlock.end))\n+ ...
nodejs/node
61,255
doc: mention --newVersion release script
This is useful when preparing a semver-minor security release
null
ac73a0799a1fe5b5735412590d02116ba3516cab
null
low
[ { "filename": "doc/contributing/releases.md", "patch": "@@ -325,7 +325,12 @@ git node release -S --prepare --security --filterLabel v20.x\n ```\n \n to automate the remaining steps until step 6 or you can perform it manually\n-following the below steps.\n+following the below steps. For semver-minors, you ca...
electron/electron
49,504
fix: avoid startup crash when V8 sandbox is disabled
Backport of #49210 See that PR for details. notes: Fixed startup crash when V8 sandbox is disabled.
null
d582f1fbaae37830658e48fcb38952a0bdfdc9bd
null
low
[ { "filename": "patches/chromium/fix_harden_blink_scriptstate_maybefrom.patch", "patch": "@@ -4,7 +4,7 @@ Date: Wed, 28 Jun 2023 21:11:40 +0900\n Subject: fix: harden blink::ScriptState::MaybeFrom\n \n NOTE: since https://chromium-review.googlesource.com/c/chromium/src/+/6973697\n-the patch is only needed fo...
ggml-org/llama.cpp
20,334
vulkan: add GATED_DELTA_NET op support
## Summary First pass at a Vulkan compute shader for `GGML_OP_GATED_DELTA_NET`, covering both the standard (scalar gate) and KDA (per-row vector gate) variants. This is the core recurrence op used by Qwen3.5 and Qwen3-Next models. **What's here:** - Autoregressive kernel: one workgroup per (head, sequence), one threa...
40c550d4f6838180fa40eda8351043353911ca40
deee23863bb75a67471c62b605248d3ff725c9f0
2
medium
[ { "filename": "ggml/src/ggml-vulkan/ggml-vulkan.cpp", "patch": "@@ -825,6 +825,8 @@ struct vk_device_struct {\n vk_pipeline pipeline_pool2d_f32;\n vk_pipeline pipeline_rwkv_wkv6_f32;\n vk_pipeline pipeline_rwkv_wkv7_f32;\n+ // [size_idx][kda] where size_idx: 0=d32, 1=d64, 2=d128\n+ vk_pipe...
huggingface/transformers
43,228
Qwen-VL video processor accepts min/max pixels
# What does this PR do? Fixes https://github.com/huggingface/transformers/issues/43183 and makes sure we can pass `min/max pixels` to video processor's call.
942f110c2fdc3a17b400b0eb8fea33580f723b5a
acb0f22679aa6229d37cec06ae9dbaa7a446bb8e
28
medium
[ { "filename": "src/transformers/models/qwen2_vl/video_processing_qwen2_vl.py", "patch": "@@ -107,6 +107,29 @@ def __init__(self, **kwargs: Unpack[Qwen2VLVideoProcessorInitKwargs]):\n \n super().__init__(size=size, min_pixels=min_pixels, max_pixels=max_pixels, **kwargs)\n \n+ def _further_process_...
facebook/react
32,119
feat(ReactNative): prioritize attribute config `process` function to allow processing function props
<!-- Thanks for submitting a pull request! We appreciate you spending the time to work on these changes. Please provide enough information so that others can review your pull request. The three fields below are mandatory. Before submitting a pull request, please make sure the following is done: 1. Fork ...
null
911dbd9e34048b21e96f24acb837b926687aa939
null
low
[ { "filename": "packages/react-native-renderer/src/ReactNativeAttributePayloadFabric.js", "patch": "@@ -254,14 +254,17 @@ function diffProperties(\n prevProp = prevProps[propKey];\n nextProp = nextProps[propKey];\n \n- // functions are converted to booleans as markers that the associated\n- // ...
facebook/react
33,481
[Flight] Use Web Streams APIs for 3rd-party component in Flight fixture
Stacked on #33480. Now that the Node bundles for Webpack also expose the Web Streams APIs, we can can use those in the Flight fixture for caching the 3rd-party component, without needing to convert the streams back and forth. This new approach of how we're exposing those APIs also allows us to mix and match them ...
b367b60927dd85239852bfee60715034c7ca97ba
c0b5a0cad32cbf237d4c0134bef702d6ba3e393c
3
medium
[ { "filename": "fixtures/flight/src/App.js", "patch": "@@ -1,6 +1,6 @@\n import * as React from 'react';\n-import {renderToPipeableStream} from 'react-server-dom-webpack/server';\n-import {createFromNodeStream} from 'react-server-dom-webpack/client';\n+import {renderToReadableStream} from 'react-server-dom-w...
vercel/next.js
88,824
[ci] Merge Cache Components and deploy tests manifests when running CC deploy tests
Part of https://linear.app/vercel/issue/NAR-719/ We previously only supported a single manifest for filtering tests. However, Cache Component deploy tests have two manifests. Now we merge manifests append-only. ## Test plan - [x] https://github.com/vercel/next.js/pull/88821 adds a failed deploy test to a test...
null
168671a57bda456a570b0e8552b413b8c9612c63
null
low
[ { "filename": ".github/workflows/build_and_test.yml", "patch": "@@ -722,7 +722,7 @@ jobs:\n afterBuild: |\n export __NEXT_CACHE_COMPONENTS=true\n export __NEXT_EXPERIMENTAL_DEBUG_CHANNEL=true\n- export NEXT_EXTERNAL_TESTS_FILTERS=\"test/cache-components-tests-manifest.json\"\n+ ...
ggml-org/llama.cpp
20,419
nit : better mtp check and fix return
cont #20411
fdb17643d379cd35bf6acf0f57cfaa500f88a145
c3e3f9e5338e8eb3d5b13d20f20d38b53d9d19ce
13
medium
[ { "filename": "convert_hf_to_gguf.py", "patch": "@@ -10092,9 +10092,9 @@ def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iter\n # Skip Multi-Token Prediction (MTP) tensors. These are used for\n # for speculative decoding but we don't include them in this m...
vuejs/vue
9,166
test(ssr): add basic directives test
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [ ] Bugf...
33e669b22f69a1f9c9147528360fe0bba85534f0
a89384c8d93a3e5a382f60a84d60f0675408ed75
24
medium
[ { "filename": "test/ssr/jasmine.js", "patch": "@@ -4,6 +4,7 @@ module.exports = {\n '*.spec.js'\n ],\n helpers: [\n- require.resolve('@babel/register')\n+ require.resolve('@babel/register'),\n+ '../helpers/to-have-been-warned.js'\n ]\n }", "additions": 2, "deletions": 1 }, { ...
ggml-org/llama.cpp
20,379
vulkan: fix SSM_CONV PP scaling with large ubatch sizes
Fixes #18725 The SSM_CONV shader dispatched one token per Y workgroup, each doing only `nc` (typically 4) multiply-adds. At ubatch=2048 this meant 2048 workgroups in Y with almost no work per launch — workgroup dispatch overhead dominated. **Changes:** - Tile 16 tokens per workgroup using a 2D local size (32×16) - Ad...
246ffc4b057504712272433b035afefc78457cda
40c550d4f6838180fa40eda8351043353911ca40
5
medium
[ { "filename": "ggml/src/ggml-vulkan/ggml-vulkan.cpp", "patch": "@@ -4576,7 +4576,7 @@ static void ggml_vk_load_shaders(vk_device& device) {\n ggml_vk_create_pipeline(device, device->pipeline_ssm_scan_f32_d256, \"ssm_scan_256_f32\", ssm_scan_f32_len, ssm_scan_f32_data, \"main\", 8, sizeof(vk_op_ssm_s...
facebook/react
33,480
[Flight] Add Web Streams APIs to unbundled Node entries for Webpack
This was missed in #33474. We're adding the new exports to the server entries only. The client entries don't need to be changed because they use `export *`.
82f3684c63fd60fdacbe4d536214596ffd7a465f
e4b88ae4c6c30791b6c1c2794d5a8e32ed19c931
7
medium
[ { "filename": "packages/react-server-dom-webpack/npm/server.node.unbundled.js", "patch": "@@ -7,9 +7,11 @@ if (process.env.NODE_ENV === 'production') {\n s = require('./cjs/react-server-dom-webpack-server.node.unbundled.development.js');\n }\n \n+exports.renderToReadableStream = s.renderToReadableStream;\...
ollama/ollama
9,484
server/.../backoff,syncs: conditionally skip synctest tests
This guards tests using the experimental synctest breaking builds in environments where the experiment is not enabled. It is required in CI, so we should still catch breaking any changes that break these packages. If we start using the new feature more, we can make it a required part of all tests.
null
55ab9f371a198a190b423763a1d48e4495fe520e
null
low
[ { "filename": "docs/development.md", "patch": "@@ -118,6 +118,35 @@ To run tests, use `go test`:\n go test ./...\n ```\n \n+> NOTE: In rare cirumstances, you may nedd to change a package using the new\n+> \"synctest\" package in go1.24.\n+>\n+> If you do not have the \"synctest\" package enabled, you will n...
facebook/react
33,438
[Flight] Clarify Semantics for Awaiting Cached Data
Technically the async call graph spans basically all the way back to the start of the app potentially, but we don't want to include everything. Similarly we don't want to include everything from previous components in every child component. So we need some heuristics for filtering out data. We roughly want to be abl...
b367b60927dd85239852bfee60715034c7ca97ba
6c8bcdaf1b0c3340150e174a342429d94e729fbb
1
medium
[ { "filename": "packages/react-client/src/__tests__/ReactFlight-test.js", "patch": "@@ -2826,46 +2826,46 @@ describe('ReactFlight', () => {\n expect(getDebugInfo(thirdPartyChildren[0])).toEqual(\n __DEV__\n ? [\n- {time: 14},\n+ {time: 22}, // Clamped to the ...
nodejs/node
58,264
worker: update code examples for `node:worker_threads` module
<!-- Before submitting a pull request, please read: - the CONTRIBUTING guide at https://github.com/nodejs/node/blob/HEAD/CONTRIBUTING.md - the commit message formatting guidelines at https://github.com/nodejs/node/blob/HEAD/doc/contributing/pull-requests.md#commit-message-guidelines For code changes: 1. Inc...
null
01e44dc9eb800e2ca5e20f724b75abe5067fe53a
null
low
[ { "filename": "doc/api/worker_threads.md", "patch": "@@ -47,8 +47,8 @@ export default function parseJSAsync(script) {\n workerData: script,\n });\n worker.on('message', resolve);\n- worker.on('error', reject);\n- worker.on('exit', (code) => {\n+ worker.once('error', reject);\n+ wor...
ggml-org/llama.cpp
20,391
CUDA: AR gated delta net improvements
I profiled the AR gated delta net, and improved perf by: 1. Adding fastdiv/fastrem for s64 int (do we even need this arithmetic to happen in 64-bit?) 2. Sharding a column across a full warp instead of using only a single thread. We don't fill SMs (at least on higher-tier GPUs) with existing launch-config (saw 16-32 C...
c6b76caf94f1da8f47ea57f17586a6e502da55b0
d1b2301f597319cea7946ed689b53e9f6288248c
2
medium
[ { "filename": "ggml/src/ggml-cuda/gated_delta_net.cu", "patch": "@@ -1,5 +1,4 @@\n #include \"gated_delta_net.cuh\"\n-#include \"ggml-cuda/common.cuh\"\n \n template <int S_v, bool KDA>\n __global__ void gated_delta_net_cuda(const float * q,\n@@ -21,15 +20,17 @@ __global__ void gated_delta_net_cuda(const fl...
huggingface/transformers
43,225
Fix failing `Hiera`, `SwiftFormer` & `LED` Model integration tests
# What does this PR do? Fixes failing [LEDModelIntegrationTests](https://github.com/huggingface/transformers/actions/runs/20888452873/job/60015690354#step:14:2651), [HieraModelIntegrationTest](https://github.com/huggingface/transformers/actions/runs/20888452873/job/60015689909#step:14:1045) & [SwiftFormer](https://g...
null
e977446e632670f9972fc4ff1432b414c8b813cb
null
low
[ { "filename": "tests/models/hiera/test_modeling_hiera.py", "patch": "@@ -557,7 +557,7 @@ def test_inference_image_classification_head(self):\n \n expected_slice = torch.tensor([[0.8028, 0.2409, -0.2254, -0.3712, -0.2848]]).to(torch_device)\n \n- torch.testing.assert_close(outputs.logits[0, :5...
electron/electron
49,444
revert: use deprecated `setAllowedFileTypes` in macOS dialogs
#### Description of Change Closes https://github.com/electron/electron/issues/48191 Fixes some issues with file type recognition owing to the new `UTType` conformance-based matching. Chrome also tried to update this and ran into the same issues and fell back. See [this comment](https://source.chromium.org/chromiu...
null
49d91dd02b96a36376f1b6ce609650feaf68d60f
null
low
[ { "filename": "shell/browser/ui/file_dialog_mac.mm", "patch": "@@ -11,7 +11,6 @@\n \n #import <Cocoa/Cocoa.h>\n #import <CoreServices/CoreServices.h>\n-#import <UniformTypeIdentifiers/UniformTypeIdentifiers.h>\n \n #include \"base/apple/foundation_util.h\"\n #include \"base/apple/scoped_cftyperef.h\"\n@@ -3...
ollama/ollama
9,468
Update README.md - Add OllamaServer
OllamaServer is an Android software that allows users to start Ollama service with one click without Termux
36dfb906bb1e28e4c2133196640ccc7424dddfd7
fefbf8f74bdedeb4a6066f0b958a52194de91240
4
medium
[ { "filename": "README.md", "patch": "@@ -513,6 +513,7 @@ See the [API documentation](./docs/api.md) for all endpoints.\n - [Maid](https://github.com/Mobile-Artificial-Intelligence/maid)\n - [Ollama App](https://github.com/JHubi1/ollama-app) (Modern and easy-to-use multi-platform client for Ollama)\n - [Conf...
vercel/next.js
88,192
Turbopack: Extend filesystem watcher fuzzing to cover symlinks and junction points
This is a crappy fuzzer I have to make sure that the filesystem watcher picks up changes to files. This didn't have coverage for links, so this tests that. I tested this on Windows and Linux. I don't care about the code quality here, so this was LLM-generated with very lightweight review.
null
d2ea7d32061c59cff57704cda4502ac794e71910
null
low
[ { "filename": "turbopack/crates/turbo-tasks-fuzz/src/main.rs", "patch": "@@ -1,3 +1,5 @@\n+#![cfg_attr(windows, feature(junction_point))]\n+\n use std::{\n fs::OpenOptions,\n io::Write,\n@@ -7,7 +9,7 @@ use std::{\n time::Duration,\n };\n \n-use clap::{Args, Parser, Subcommand};\n+use clap::{Arg...
facebook/react
33,473
[Flight] Add Web Stream support to the Flight Client in Node
This effectively lets us consume Web Streams in a Node build. In fact the Node entry point is now just adding Node stream APIs. For the client, this is simple because the configs are not actually stream type specific. The server is a little trickier.
82f3684c63fd60fdacbe4d536214596ffd7a465f
280ff6fed2a84b6ad7588c72d3e66b20f0f3c91a
1
medium
[ { "filename": "packages/react-server-dom-parcel/src/client/ReactFlightDOMClientNode.js", "patch": "@@ -19,9 +19,7 @@ import {\n close,\n } from 'react-client/src/ReactFlightClient';\n \n-import {createServerReference as createServerReferenceImpl} from 'react-client/src/ReactFlightReplyClient';\n-\n-export...
vuejs/vue
8,381
fix(lifecycle): updated should not be called after component being de…
…stroyed, fix #8076 <!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at l...
null
a64ff1957c35270b818aa9cfdfb2acb6e42ce2a9
null
low
[ { "filename": "src/core/observer/scheduler.js", "patch": "@@ -98,7 +98,7 @@ function callUpdatedHooks (queue) {\n while (i--) {\n const watcher = queue[i]\n const vm = watcher.vm\n- if (vm._watcher === watcher && vm._isMounted) {\n+ if (vm._watcher === watcher && vm._isMounted && !vm._isDest...
huggingface/transformers
43,259
fix `_get_test_info` in `testing_utils.py`
# What does this PR do? This method uses some check like > frame.function == test_name however, when we use `@parameterized.expand`, - `frame.function` could be `test_model_logits` - `test_name` could be `test_model_logits_0_cpu` and we got some strange error message. See ``` (Pdb) frame.functi...
5f70e191f8710c21e358af9ff6cf225b94fed36f
d79b0b885288cc71876b5d34e2c315bf66265f2a
21
medium
[ { "filename": "src/transformers/testing_utils.py", "patch": "@@ -3318,7 +3318,7 @@ def _get_test_info():\n # check frame's function + if it has `self` as locals; double check if self has the (function) name\n # TODO: Question: How about expanded?\n if (\n- frame.function =...
ggml-org/llama.cpp
20,341
[ggml-virtgpu] Fix some build commands of `development.md`
Hi, this PR fixes the bug of build command in the document. - Fix incorrect cmake option in building host `ggml-virtgpu` (`GGML_REMOTINGBACKEND=ONLY` → `GGML_VIRTGPU=ON` + `GGML_VIRTGPU_BACKEND=ONLY`) - Add missing `cd virglrenderer` step in building virglrenderer - Fix double slash typo in `$PWD//build-virtgpu`
null
05039967da783fa8c7b8bc51e80ae834ab1cda73
null
low
[ { "filename": "docs/backend/VirtGPU/development.md", "patch": "@@ -55,7 +55,8 @@ LLAMA_MAC_BUILD=$PWD/build/ggml-virtgpu-backend\n cmake -S . -B $LLAMA_MAC_BUILD \\\n -DGGML_NATIVE=OFF \\\n -DLLAMA_CURL=ON \\\n- -DGGML_REMOTINGBACKEND=ONLY \\\n+ -DGGML_VIRTGPU=ON \\\n+ -DGGML_VIRT...
facebook/react
33,467
[Fizz] Support deeply nested Suspense inside fallback
When deeply nested Suspense boundaries inside a fallback of another boundary resolve it is possible to encounter situations where you either attempt to flush an aborted Segment or you have a boundary without any root segment. We intended for both of these conditions to be impossible to arrive at legitimately however it...
d177272802b7f86a847312c23b7e60a6f56434de
142aa0744d0e73dc5390bc19d4d41dd8aeda2b19
5
medium
[ { "filename": "packages/react-dom/src/__tests__/ReactDOMFizzServer-test.js", "patch": "@@ -88,6 +88,7 @@ describe('ReactDOMFizzServer', () => {\n setTimeout(cb);\n container = document.getElementById('container');\n \n+ CSPnonce = null;\n Scheduler = require('scheduler');\n React = requ...
nodejs/node
61,260
doc: correct typo in api contributing doc
In [doc/contributing/api-documentation.md#vocabulary--good-to-knows](https://github.com/nodejs/node/blob/main/doc/contributing/api-documentation.md#vocabulary--good-to-knows) and viewing the sentence: > * The [Stability Index](https://nodejs.org/dist/latest/docs/api/documentation.html#stability-index) is used to com...
null
5246229a895d40046f9d4fe3af675d02e776b9de
null
low
[ { "filename": "doc/contributing/api-documentation.md", "patch": "@@ -54,7 +54,7 @@ meant to be a guide on how to write documentation for Node.js.\n it as a reference for the structure of the documentation.\n * The\n [Stability Index](https://nodejs.org/dist/latest/docs/api/documentation.html#stability-i...
ollama/ollama
9,481
fix(install.sh): update lib/ollama directory permissions
this change explicitly creates the `lib/ollama` directory with the correct permissions. since the ollama release tarball does not contain the `lib/ollama` entry, it's implicitly created when extracting the archive. this can result in a base directory without other read/execute and prevent ollama from accessing the libr...
null
ba7d31240eb0b44064a8220a8a6143307b1459ef
null
low
[ { "filename": "llama/patches/0015-try-catch-backend-load.patch", "patch": "@@ -1,69 +0,0 @@\n-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001\n-From: Michael Yang <mxyng@pm.me>\n-Date: Tue, 11 Feb 2025 14:06:36 -0800\n-Subject: [PATCH] try/catch backend load\n-\n----\n- ggml/src/ggml-...
electron/electron
49,471
revert: use deprecated `setAllowedFileTypes` in macOS dialogs
Backport of #49444 See that PR for details. Notes: Fixed an issue where some packages weren't correctly filtered on macOS in dialogs.
null
8fb777dab01b13925721831a2934c20a4414f18d
null
low
[ { "filename": "shell/browser/ui/file_dialog_mac.mm", "patch": "@@ -11,7 +11,6 @@\n \n #import <Cocoa/Cocoa.h>\n #import <CoreServices/CoreServices.h>\n-#import <UniformTypeIdentifiers/UniformTypeIdentifiers.h>\n \n #include \"base/apple/foundation_util.h\"\n #include \"base/apple/scoped_cftyperef.h\"\n@@ -3...
facebook/react
33,463
[Fizz] Shorten throttle to hit a specific target metric
Adding throttling or delaying on images, can obviously impact metrics. However, it's all in the name of better actual user experience overall. (Note that it's not strictly worse even for metric. Often it's actually strictly better due to less work being done overall thanks to batching.) Metrics can impact things lik...
22b929156c325eaf52c375f0c62801831951814a
6ccf328499f06c140ffe96a096744c22319394cc
5
medium
[ { "filename": "packages/react-dom-bindings/src/server/fizz-instruction-set/ReactDOMFizzInstructionSetInlineCodeStrings.js", "patch": "@@ -6,9 +6,9 @@ export const markShellTime =\n export const clientRenderBoundary =\n '$RX=function(b,c,d,e,f){var a=document.getElementById(b);a&&(b=a.previousSibling,b.dat...
ollama/ollama
9,479
docker: use go version from go.mod
use the go.mod version so it's always up to date. this skips downloading a compatible version when `ARG GOVERSION` is out of date
1579c4f06d7c6cc96a91f76be3e0dcad7839e9eb
b428ddd79686a0c72db51f55fe0673e47cbf094e
6
medium
[ { "filename": "Dockerfile", "patch": "@@ -86,10 +86,11 @@ RUN --mount=type=cache,target=/root/.ccache \\\n && cmake --install build --component CUDA --strip --parallel 8\n \n FROM base AS build\n-ARG GOVERSION=1.23.4\n-RUN curl -fsSL https://golang.org/dl/go${GOVERSION}.linux-$(case $(uname -m) in x...
vercel/next.js
88,829
[test]: add ability to run test-deploy with pre-existing deployment
When digging into failing deployment tests, I often will re-run the test itself without wanting to change anything about Next.js or the test application, especially when I suspect a logic bug in the test itself or need to figure out what's causing a flake. This adds an environment variable that can be used to skip t...
null
6138452203337a72d79decbe26d7be0bd36233f4
null
low
[ { "filename": "contributing/core/testing.md", "patch": "@@ -92,6 +92,8 @@ these can be leveraged by prefixing the `pnpm test` command.\n it can be used when not using `pnpm test-dev` or `pnpm test-start` directly.\n Valid test modes can be seen here:\n https://github.com/vercel/next.js/blob/aa664868c1...
vuejs/vue
8,438
types: add AsyncComponentFactory type
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [ ] Bugf...
f43ae6d151de3ab00a5fbd1bb5fd3fabf454de85
dfaf9e24361e10ae68ce3951eaf48262cf90f0ec
22
medium
[ { "filename": "types/options.d.ts", "patch": "@@ -15,12 +15,24 @@ export type Component<Data=DefaultData<never>, Methods=DefaultMethods<never>, Co\n interface EsModuleComponent {\n default: Component\n }\n+ \n+export type AsyncComponent<Data=DefaultData<never>, Methods=DefaultMethods<...
ollama/ollama
9,424
Update README.md
removed phi 3 mini and added phi4-mini
3b1ddb2b3a00598162a69b0441e4e56df3af7666
a6f0f908b93c05319fb58b682c2b2123c160af3d
1
high
[ { "filename": "README.md", "patch": "@@ -64,7 +64,7 @@ Here are some example models that can be downloaded:\n | Llama 3.1 | 8B | 4.7GB | `ollama run llama3.1` |\n | Llama 3.1 | 405B | 231GB | `ollama run llama3.1:405b` |\n | Phi 4 | 14B | ...
vuejs/vue
9,150
fix #9112: fix keyName checking for space and delete in IE11
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [x] Bugf...
1b4a8a0c1edaf9c7eb129ba61bca94ba607bbf56
0ed0aad77228b95e9a61a87386736938837527f8
3
medium
[ { "filename": "src/compiler/codegen/events.js", "patch": "@@ -22,13 +22,15 @@ const keyNames: { [key: string]: string | Array<string> } = {\n esc: ['Esc', 'Escape'],\n tab: 'Tab',\n enter: 'Enter',\n- space: ' ',\n+ // #9112: IE11 uses `Spacebar` for Space key name.\n+ space: [' ', 'Spacebar'],\n ...
ollama/ollama
9,475
build: install binutils with gcc 10
e41c4cbea722e40f8e346e5a6c4344a8f47d06d9
1579c4f06d7c6cc96a91f76be3e0dcad7839e9eb
2
high
[ { "filename": "Dockerfile", "patch": "@@ -12,9 +12,8 @@ FROM --platform=linux/amd64 rocm/dev-almalinux-8:${ROCMVERSION}-complete AS base\n RUN yum install -y yum-utils \\\n && yum-config-manager --add-repo https://dl.rockylinux.org/vault/rocky/8.5/AppStream/\\$basearch/os/ \\\n && rpm --import https...
ggml-org/llama.cpp
20,403
New conversations now auto-select the first loaded model
In router mode, new conversations now auto-select the first loaded model instead of showing an empty dropdown. Existing conversations still pick up the last model used in each conversation when switching between chats. Solves #20382
null
de190154c85d20e24dbeae8c8af1849402ae5098
null
low
[ { "filename": "tools/server/webui/src/lib/components/app/chat/ChatForm/ChatFormActions/ChatFormActions.svelte", "patch": "@@ -62,15 +62,12 @@\n \t\tchatStore.getConversationModel(activeMessages() as DatabaseMessage[])\n \t);\n \n-\tlet previousConversationModel: string | null = null;\n-\n \t$effect(() => {\...
huggingface/transformers
43,252
Fix failing `Vip-llava` model integration test
# What does this PR do? Fixes this failing [VipLlavaForConditionalGenerationIntegrationTest](https://github.com/huggingface/transformers/actions/runs/20888452873/job/60015689982#step:14:2081) <img width="1324" height="187" alt="image" src="https://github.com/user-attachments/assets/7553a419-54fc-414d-966e-316fe5a...
null
ad7f4d0103599ff098bb33c11b9c1a73d97262fd
null
low
[ { "filename": "tests/models/vipllava/test_modeling_vipllava.py", "patch": "@@ -303,8 +303,7 @@ def test_small_model_integration_test(self):\n image = Image.open(requests.get(url, stream=True).raw)\n prompt = \"USER: <image>\\nCan you please describe this image?\\nASSISTANT:\"\n \n- in...
vercel/next.js
88,751
Rename `rewroteURL` to `rewrittenPathname` in request metadata
The field stores a pathname, not a full URL.
b58b1dc901a2deed30e6861e0675352c92dd925e
06421057b5d261388e23f31f1666de051ee0e2c6
7
medium
[ { "filename": "packages/next/src/server/app-render/app-render.tsx", "patch": "@@ -2023,9 +2023,8 @@ async function renderToHTMLOrFlightImpl(\n // fetch cache. Using the destination path ensures that\n // revalidatePath('/dest') invalidates cache entries for pages rewritten to\n // that destination.\n-...
nodejs/node
60,511
build: fix inconsistent quoting in `Makefile`
Currently `available-node` results in something that looks like `if [ -x ""…/node"" ] && [ -e ""…/node"" ]; then…`, the double quoting is clearly not helpful. I went with using single quotes in case the path contains weird chars (although it wouldn't help if PWD itself contains a single quote, but oh well), I left out...
null
6218d14505668002ece19dd91b3349e4d3967759
null
low
[ { "filename": "Makefile", "patch": "@@ -17,7 +17,7 @@ endif\n ARCHTYPE := $(shell uname -m | tr '[:upper:]' '[:lower:]')\n COVTESTS ?= test-cov\n COV_SKIP_TESTS ?= core_line_numbers.js,testFinalizer.js,test_function/test.js\n-GTEST_FILTER ?= \"*\"\n+GTEST_FILTER ?= *\n GNUMAKEFLAGS += --no-print-directory\n...
electron/electron
49,346
fix: restore AXDocument accessibility attribute for representedFilename on macOS
#### Description of Change This PR fixes a regression where `setRepresentedFilename()` no longer sets the `AXDocument` accessibility attribute on macOS windows starting from Electron 35.0.0-beta.5. ## Root Cause Chromium change [crrev.com/c/6187085](https://chromium-review.googlesource.com/c/chromium/src/+/618...
null
5dd509d2f4ec6a6e862d8f845a3ab95578a5631b
null
low
[ { "filename": "shell/browser/ui/cocoa/electron_ns_window.mm", "patch": "@@ -285,6 +285,19 @@ - (NSString*)accessibilityTitle {\n return base::SysUTF8ToNSString(shell_ ? shell_->GetTitle() : \"\");\n }\n \n+- (NSString*)accessibilityDocument {\n+ // Prefer representedFilename set via Electron's setReprese...
facebook/react
33,466
[ci] Fix missing permissions for stale job
Missed these the last time.
null
a374e0ec87ec1d45a94b69e26c747529ea5dbab0
null
low
[ { "filename": ".github/workflows/shared_stale.yml", "patch": "@@ -6,7 +6,10 @@ on:\n - cron: '0 * * * *'\n workflow_dispatch:\n \n-permissions: {}\n+permissions:\n+ # https://github.com/actions/stale/tree/v9/?tab=readme-ov-file#recommended-permissions\n+ issues: write\n+ pull-requests: write\n \n e...
ggml-org/llama.cpp
20,426
metal : avoid divisions in bin kernel
Before: <img width="1117" height="269" alt="image" src="https://github.com/user-attachments/assets/cc9e955a-c5df-423a-bf81-6dd5cb8fd4f1" /> <img width="951" height="519" alt="image" src="https://github.com/user-attachments/assets/ed7907b2-4935-4d00-b0fb-56c3ad7a9b88" /> After: <img width="1126" height="27...
d28961d81e73e32b295d0ad638f3ff14676aeeda
e4cff0956b0dd7196cd8d4717daf02329cdeb870
13
medium
[ { "filename": "ggml/src/ggml-metal/ggml-metal-context.m", "patch": "@@ -554,7 +554,7 @@ enum ggml_status ggml_metal_graph_compute(ggml_metal_t ctx, struct ggml_cgraph *\n \n // enter here only when capturing in order to wait for all computation to finish\n // otherwise, we leave the graph to...
vercel/next.js
88,733
Only filter next config if experimental flag is enabled
This filtering would be a breaking change: https://github.com/vercel/next.js/pull/86812
7e9d7cbd60325bf916af6e0d47c81d2071867386
38f4304c2255b3c6443796a052fc3ed0f8cb3aa4
13
medium
[ { "filename": "packages/next/src/build/index.ts", "patch": "@@ -1904,11 +1904,15 @@ export default async function build(\n }\n }\n \n+ // getNextConfigRuntime only filters if experimental.runtimeServerDeploymentId is true,\n+ // but we unconditionally want to remove c...
vuejs/vue
9,131
fix(types): correct scopedSlot types
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [x] Bugf...
76443803e8878dc13a31ebed119abfd31573e7f1
448ba65d2b139b29f1e6891add9925ac22ffe10b
20
medium
[ { "filename": "types/test/options-test.ts", "patch": "@@ -1,5 +1,5 @@\n import Vue, { VNode } from \"../index\";\n-import { AsyncComponent, ComponentOptions, FunctionalComponentOptions, Component } from \"../index\";\n+import { ComponentOptions, Component } from \"../index\";\n import { CreateElement } from...
ggml-org/llama.cpp
20,296
vulkan: fix OOB check in flash_attn_mask_opt
Fixes #19955. I saw a few percent slowdown with pp512 (which is too small to hit the aligned path on my system after this change) so I tweaked the use_mask_opt logic to hide it. I should look into spreading the work across more workgroups, but I don't have time for that today. @el95149 this is different enough fr...
5866e3bbc8a2ede4236871ea73a3eb3337abf530
aa429cf5075ef1cf14c9cf784c73fc436ff6eede
1
medium
[ { "filename": "ggml/src/ggml-vulkan/ggml-vulkan.cpp", "patch": "@@ -8840,7 +8840,7 @@ static void ggml_vk_flash_attn(ggml_backend_vk_context * ctx, vk_context& subctx\n }\n \n // Only use mask opt when the mask is fairly large. This hasn't been tuned extensively.\n- bool use_mask_opt = mask && ne...
huggingface/transformers
43,250
Adding to run slow
# What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this w...
5f70e191f8710c21e358af9ff6cf225b94fed36f
35c26ace56c3d0190c0aa3e3dcefdc67c79abcf9
18
medium
[ { "filename": ".github/workflows/pr-repo-consistency-bot.yml", "patch": "@@ -15,7 +15,7 @@ permissions: read-all\n jobs:\n get-pr-number:\n name: Get PR number\n- if: ${{ github.event.issue.state == 'open' && contains(fromJSON('[\"ydshieh\", \"ArthurZucker\", \"zucchini-nlp\", \"molbap\", \"gante\"...
huggingface/transformers
43,253
Fix benchmark script
# What does this PR do? Fix after https://github.com/huggingface/transformers/pull/43153. F821 Undefined name `device_type` which crashes `make style` etc cc @remi-or for viz!
3ef2e3ba491c14b6b9bf1d2c19735781078a7127
b40b7a767d569b067a115ce115fd270ac323cd36
1
medium
[ { "filename": "benchmark_v2/framework/benchmark_config.py", "patch": "@@ -5,11 +5,12 @@\n from functools import lru_cache\n from typing import Any\n \n+import torch\n+\n from transformers.generation.configuration_utils import CompileConfig\n-from transformers.utils.import_utils import is_flash_attn_2_availa...
vuejs/vue
8,946
Typings: Improve $slots and $scopedSlots type to prevent unchecked access to undefined
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [x] Bugf...
3cd4af4af0a8a67f5887d5fc967147d433c8612c
76443803e8878dc13a31ebed119abfd31573e7f1
1
medium
[ { "filename": "types/test/options-test.ts", "patch": "@@ -304,8 +304,8 @@ Vue.component('component-with-scoped-slot', {\n child: {\n render (this: Vue, h: CreateElement) {\n return h('div', [\n- this.$scopedSlots['default']({ msg: 'hi' }),\n- this.$scopedSlots['item']({ m...
ollama/ollama
9,464
build: install ccache manually in Dockerfile
Reverts ccache installation to be done manually via curl instead of using the dnf package manager as this has side effects of prepending ccache's install directory to the front of the PATH
null
e41c4cbea722e40f8e346e5a6c4344a8f47d06d9
null
low
[ { "filename": "Dockerfile", "patch": "@@ -12,8 +12,9 @@ FROM --platform=linux/amd64 rocm/dev-almalinux-8:${ROCMVERSION}-complete AS base\n RUN yum install -y yum-utils \\\n && yum-config-manager --add-repo https://dl.rockylinux.org/vault/rocky/8.5/AppStream/\\$basearch/os/ \\\n && rpm --import https...
nodejs/node
61,256
doc: add PR-URL requirement for security backports
Security release automation requires that all backport commits include PR-URL metadata in their commit messages. Refs: https://github.com/nodejs/node-core-utils/pull/1022
null
088e63842a82f3531220613f1bec60b789f67a11
null
low
[ { "filename": "doc/contributing/security-release-process.md", "patch": "@@ -83,6 +83,8 @@ The current security stewards are documented in the main Node.js\n * Get volunteers for the upcoming security release on the affected release\n lines.\n * Make sure to sync nodejs-private (vN.x) branches with n...
huggingface/transformers
42,880
fix: make mistral base check conditional to fix offline loading
Backport of a small part of #42528, the model_info call currently breaks offline loading
d3ee5e8f146ef9fb7299a61a5e406b49bf6b460c
33ac6bd3908dd20d2b16c12b1acbe4a003115243
2
medium
[ { "filename": "src/transformers/tokenization_utils_base.py", "patch": "@@ -2435,7 +2435,12 @@ def is_base_mistral(model_id: str) -> bool:\n return True\n return False\n \n- if _is_local or is_base_mistral(pretrained_model_name_or_path):\n+ if is_offline_mode():\...
ollama/ollama
9,454
server/internal/client/ollama: handle extended names in client/ollama
The extended name format is a superset of the name format that only the client needs to know about, not the server or other dependents of the name package, so move the split logic into the client package.
cda6f5c66c13b79a3dc9fb172ab5b17da9ce1bed
ee048b76d42fbfb108701682cf71c57daeb6678b
8
medium
[ { "filename": "server/internal/client/ollama/registry.go", "patch": "@@ -212,12 +212,16 @@ type Registry struct {\n \tMask string\n }\n \n-func (r *Registry) completeName(name string) names.Name {\n+func (r *Registry) parseName(name string) (names.Name, error) {\n \tmask := defaultMask\n \tif r.Mask != \"\"...
facebook/react
33,433
[Fizz] Suspensey Images for View Transition Reveals
Block the view transition on suspensey images Up to 500ms just like the client. We can't use `decode()` because a bug in Chrome where those are blocked on `startViewTransition` finishing we instead rely on sync decoding but also that the image is live when it's animating in and we assume it doesn't start visible. ...
1ae0a845bde5b95dfc319cadf366cb7b3fb1ca92
22b929156c325eaf52c375f0c62801831951814a
19
medium
[ { "filename": "packages/react-dom-bindings/src/server/ReactFizzConfigDOM.js", "patch": "@@ -4860,8 +4860,9 @@ export function writeCompletedSegmentInstruction(\n const completeBoundaryScriptFunctionOnly = stringToPrecomputedChunk(\n completeBoundaryFunction,\n );\n-const completeBoundaryUpgradeToViewTrans...
ggml-org/llama.cpp
20,154
ci: Setup self-hosted CI for Intel Linux Vulkan backend
We would like add a single self-hosted CI runner for Intel Linux Vulkan backend related to #19213. This will be focused for Vulkan backend only at the moment, but when OpenVINO backend gets merged we will use it for both backends. An example workflow execution on the branch can be viewed at the following: - https...
182acfe5c5eb17a4f82d9181fa7bd91510e3b93d
4cc6eb158c83a3b9833526f3bf1175696884fb0c
22
medium
[ { "filename": ".github/workflows/build.yml", "patch": "@@ -1727,6 +1727,22 @@ jobs:\n vulkaninfo --summary\n GG_BUILD_VULKAN=1 bash ./ci/run.sh ~/results/llama.cpp ~/mnt/llama.cpp\n \n+ ggml-ci-x64-linux-intel-vulkan:\n+ runs-on: [self-hosted, Linux, X64, Intel]\n+\n+ steps:\n+ ...
electron/electron
49,484
fix: return early from platform_util::Beep() on Linux if there is no default GDK display
Backport of #49442 See that PR for details. Notes: Fixed crash in platform_util::Beep() on Linux
null
06fdee87b3fb2ce7a5ad730b2397bee00e289f1d
null
low
[ { "filename": "shell/common/platform_util_linux.cc", "patch": "@@ -406,6 +406,8 @@ bool PlatformTrashItem(const base::FilePath& full_path, std::string* error) {\n \n void Beep() {\n auto* display = gdk_display_get_default();\n+ if (!display)\n+ return;\n gdk_display_beep(display);\n }\n ", "addi...
ggml-org/llama.cpp
20,350
vulkan: fix l2_norm epsilon handling
This is the only "real" bug I could find in test-llama-archs. I see some other failures but they may be driver/compiler bugs.
aa429cf5075ef1cf14c9cf784c73fc436ff6eede
246ffc4b057504712272433b035afefc78457cda
1
medium
[ { "filename": "ggml/src/ggml-vulkan/ggml-vulkan.cpp", "patch": "@@ -16061,7 +16061,7 @@ static void ggml_vk_check_results_0(ggml_backend_vk_context * ctx, ggml_cgraph *\n tensor_clone = ggml_arange(ggml_ctx, start, stop, step);\n } else if (tensor->op == GGML_OP_FILL) {\n con...
vercel/next.js
88,664
[Devtool Indicator] Fix cross alignment
<img width="4106" height="3504" alt="image" src="https://github.com/user-attachments/assets/102ddcd6-84c3-4cf7-be99-977473399570" />
null
a6f714cbc9cf6f4538cc61385d81a3f1cb90c82e
null
low
[ { "filename": "packages/next/src/next-devtools/dev-overlay/components/devtools-indicator/next-logo.tsx", "patch": "@@ -237,10 +237,6 @@ export function NextLogo({\n &:has([data-issues-collapse]) {\n padding-right: calc(var(--padding-left) / 2);\n }\n-\n- [dat...
vuejs/vue
9,129
fix #8961 resolve camelCased local directives
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [ ] Bugf...
null
307835284a326569ea12c4a22c7dcb8f36d2d8ca
null
low
[ { "filename": "src/server/render.js", "patch": "@@ -3,6 +3,7 @@\n import { escape } from 'web/server/util'\n import { SSR_ATTR } from 'shared/constants'\n import { RenderContext } from './render-context'\n+import { resolveAsset } from 'core/util/options'\n import { generateComponentTrace } from 'core/util/d...
ollama/ollama
9,442
Add AstrBot to Community Integrations
Hi, AstrBot is a user-friendly LLM-based multi-platform chatbot with a WebUI, and it supports RAG features, LLM agents, and plugins integration. Opensource Repo: https://github.com/Soulter/AstrBot Related docs: https://astrbot.app/config/providers/provider-ollama.html
25885e5335d85c615678bbb683fdc7157c40b9bd
af68d60a58c7ade657a1f4641bf301a29d977174
14
medium
[ { "filename": "README.md", "patch": "@@ -387,6 +387,7 @@ See the [API documentation](./docs/api.md) for all endpoints.\n - [yla](https://github.com/danielekp/yla) (Web interface to freely interact with your customized models)\n - [LangBot](https://github.com/RockChinQ/LangBot) (LLM-based instant messaging b...
huggingface/transformers
42,990
Deprecate dtype per sub config
# What does this PR do? Fixes https://github.com/huggingface/transformers/issues/42968, https://github.com/huggingface/transformers/issues/42641 As discussed, this PR deprecated different dtypes per backbone and falls back to the main dtype if a user is requesting different dtypes. There is no clean way to su...
5f70e191f8710c21e358af9ff6cf225b94fed36f
7918b152c83d2f93f28869f2a773b9b8bc2a5650
11
medium
[ { "filename": "src/transformers/modeling_utils.py", "patch": "@@ -759,7 +759,6 @@ def _get_dtype(\n 2. Else, use the dtype provided as a dict or str\n \"\"\"\n is_sharded = sharded_metadata is not None\n- asked_dtype = dtype\n \n if dtype is not None:\n if isinstance(dtype, str):\...
vercel/next.js
88,744
Allow inspecting server with `next start --inspect`
Same as https://github.com/vercel/next.js/pull/85037 but for `next start` `NODE_OPTIONS=--inspect` is way too convoluted with package.json scripts. This implements `--inspect` for `next start` which will forward the value for the Node.js process of the server (which is where user code will be executed). `NODE_O...
b46d599b0fe6bae102f98dac49cfeec0f1102eb6
d56ed7bbca98107d76293a69293770cdc80f98e3
4
medium
[ { "filename": "docs/01-app/02-guides/debugging.mdx", "patch": "@@ -115,13 +115,16 @@ To debug server-side Next.js code with browser DevTools, you need to pass the `-\n \n ```bash filename=\"Terminal\"\n next dev --inspect\n+# or\n+next start --inspect\n+# next build --inspect is currently not supported\n ``...
ollama/ollama
9,383
attention: Remove unnecessary contiguous operations & Flash attention
This has two significant performance improvements related to attention on the new engine: - Removing extra contiguous calls after permutations required before attention. This improves token generation performance by nearly 25%. - Implement support for flash attention, which improves performance by roughly another 10%...
ee141cc82174ff235c9f26f661c7f9d2cd36d312
21aa666a1eeff87d3fc6f4f8a43167e1fdd0d3ad
1
medium
[ { "filename": "kvcache/causal.go", "patch": "@@ -90,6 +90,14 @@ func (c *Causal) Init(backend ml.Backend, dtype ml.DType, capacity int32) {\n \t\tc.config.CachePadding = 1\n \t}\n \n+\tif c.config.MaskBatchPadding == 0 {\n+\t\tc.config.MaskBatchPadding = 1\n+\t}\n+\n+\tif c.config.MaskDType == ml.DTypeOther...
facebook/react
33,452
[tests] remove pretest compiler script
This shouldn't be needed now that the lint rule was move
3531b2672955285c0a59f84be06ac92b0d987a3e
a3be6829c6425f306a8bef9f7dba72d1347a64b3
20
medium
[ { "filename": ".github/workflows/runtime_build_and_test.yml", "patch": "@@ -280,6 +280,35 @@ jobs:\n if: steps.node_modules.outputs.cache-hit != 'true'\n - run: yarn test ${{ matrix.params }} --ci --shard=${{ matrix.shard }}\n \n+ # Hardcoded to improve parallelism\n+ test-linter:\n+ name...
ollama/ollama
9,451
build: use correct GGML_HIP_NO_VMM definition for ggml-hip
e75c6126e92e592440b0b6b0966bcf4a0868e05a
96a97adf9b973721f1e7401e49984cc1b772cc3b
1
high
[ { "filename": "CMakeLists.txt", "patch": "@@ -109,7 +109,7 @@ if(CMAKE_HIP_COMPILER)\n target_compile_definitions(ggml-hip PRIVATE GGML_CUDA_NO_PEER_COPY)\n endif()\n \n- target_compile_definitions(ggml-hip PRIVATE GGML_CUDA_NO_VMM)\n+ target_compile_definitions(ggml-hip PR...
facebook/react
33,445
[Flight] Bypass caches in Flight fixture if requested
When the "Disable cache" checkbox is checked in the dev tools, we should bypass the caching of the 3rd-party component in the Flight fixture. This allows us to quickly compare the different output (with and without caching) in React's custom performance tracks.
b4477d3800ccb0bdf26670cd1f021d094159c38f
b1759882c0b8045aff27fa9e41600534d396f69c
4
medium
[ { "filename": "fixtures/flight/server/global.js", "patch": "@@ -101,6 +101,9 @@ async function renderApp(req, res, next) {\n } else if (req.get('Content-type')) {\n proxiedHeaders['Content-type'] = req.get('Content-type');\n }\n+ if (req.headers['cache-control']) {\n+ proxiedHeaders['Cache-Contr...
nodejs/node
56,628
module: add dynamic file-specific ESM warnings
I improved the error message to include specific file and package.json details, trying to make it clearer and more actionable as requested in the TODO
null
7c910d09e29186775866d1b69fa1be1d9e900d0a
null
low
[ { "filename": "src/node_contextify.cc", "patch": "@@ -1702,13 +1702,19 @@ static MaybeLocal<Function> CompileFunctionForCJSLoader(\n return scope.Escape(fn);\n }\n \n+static std::string GetRequireEsmWarning(Local<String> filename) {\n+ Isolate* isolate = Isolate::GetCurrent();\n+ Utf8Value filename_utf8...
ggml-org/llama.cpp
20,059
vulkan: Fix ErrorOutOfHostMemory on Intel GPU when loading large models with --no-mmap
Fixes #19420. # Overview We were hitting an internal maximum number (16383) of command buffers for Intel's Windows GPU driver causing ErrorOutOfHostMemory when loading large models (1MB per transfer * 16383 == approx 16GB or more weight). This PR attempts to fix this by reusing command buffers that are done transfe...
null
5866e3bbc8a2ede4236871ea73a3eb3337abf530
null
low
[ { "filename": "ggml/src/ggml-vulkan/ggml-vulkan.cpp", "patch": "@@ -27,6 +27,7 @@ DispatchLoaderDynamic & ggml_vk_default_dispatcher();\n #include <iostream>\n #include <tuple>\n #include <vector>\n+#include <deque>\n #include <sstream>\n #include <utility>\n #include <memory>\n@@ -188,17 +189,28 @@ struct ...
ollama/ollama
9,449
build: set GGML_CUDA_NO_VMM=1 for ROCm builds
b42aba40ed21862532b9d195a28915a1fca88560
e75c6126e92e592440b0b6b0966bcf4a0868e05a
7
medium
[ { "filename": "CMakeLists.txt", "patch": "@@ -106,9 +106,11 @@ if(CMAKE_HIP_COMPILER)\n add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/ml/backend/ggml/ggml/src/ggml-hip)\n \n if (WIN32)\n- target_compile_definitions(ggml-hip PRIVATE GGML_CUDA_NO_PEER_COPY=1)\n+ target_com...
vuejs/vue
8,608
ci: add regression test for popular libraries in Vue.js ecosystem
See https://github.com/vuejs/regression-testing **What kind of change does this PR introduce?** (check at least one) - [ ] Bugfix - [ ] Feature - [ ] Code style update - [ ] Refactor - [x] Build-related changes - [ ] Other, please describe: **Does this PR introduce a breaking change?** (check one) - [ ...
null
aca17b4064bf5017506d77e9f5bf286cfe926ae1
null
low
[ { "filename": ".circleci/config.yml", "patch": "@@ -62,6 +62,16 @@ jobs:\n - run: npm run test:ssr\n - run: npm run test:weex\n \n+ trigger-regression-test:\n+ <<: *defaults\n+ steps:\n+ - run:\n+ command: |\n+ curl --user ${CIRCLE_TOKEN}: \\\n+ --data ...
ggml-org/llama.cpp
20,316
opencl: use larger workgroup size for get_rows
Choose the largest possible workgroup size for get_rows for better performance plus small refactor.
3d9ab225e71d96458adc2219f0d039bc46dbc8b9
0516e04bf9c002bf190519af6795415ab3ebd059
1
medium
[ { "filename": "ggml/src/ggml-opencl/ggml-opencl.cpp", "patch": "@@ -5796,19 +5796,12 @@ static void ggml_cl_get_rows(ggml_backend_t backend, const ggml_tensor * src0, c\n GGML_ASSERT(dst);\n GGML_ASSERT(dst->extra);\n \n- const int ne00 = src0->ne[0];\n- const cl_ulong nb01 = src0->nb[1];...
vercel/next.js
88,754
[test] Deflake `test/development/app-dir/browser-log-forwarding/fixtures/verbose-level/verbose-level.test.ts`
Fixes ``` browser-log-forwarding verbose level › should forward all logs to terminal expect(received).toMatchInlineSnapshot(snapshot) Snapshot name: `browser-log-forwarding verbose level should forward all logs to terminal 1` - Snapshot - 1 + Received + 0 [browser] browser log: this is ...
8cfdcacf376cb8088247ba870004dbd2787078cd
a0e2d59cf476981d18adcafa912d014664cd879a
26
medium
[ { "filename": "test/development/app-dir/browser-log-forwarding/fixtures/verbose-level/verbose-level.test.ts", "patch": "@@ -15,6 +15,7 @@ describe('browser-log-forwarding verbose level', () => {\n expect(output).toContain('browser error:')\n expect(output).toContain('browser warn:')\n expe...
electron/electron
49,483
fix: return early from platform_util::Beep() on Linux if there is no default GDK display
Backport of #49442 See that PR for details. Notes: Fixed crash in platform_util::Beep() on Linux
null
173d0d16dc5e6aa50f2d12aadbe99108a199909a
null
low
[ { "filename": "shell/common/platform_util_linux.cc", "patch": "@@ -406,6 +406,8 @@ bool PlatformTrashItem(const base::FilePath& full_path, std::string* error) {\n \n void Beep() {\n auto* display = gdk_display_get_default();\n+ if (!display)\n+ return;\n gdk_display_beep(display);\n }\n ", "addi...
huggingface/transformers
43,140
Fix unsafe torch.load() in _load_rng_state allowing arbitrary code execution
Add weights_only=True to torch.load() call in Trainer._load_rng_state() to prevent arbitrary code execution when loading malicious checkpoint files. The existing safe_globals() context manager provides no protection for PyTorch < 2.6 as it returns contextlib.nullcontext(). This makes the torch.load() call at line 30...
null
03c8082ba4594c9b8d6fe190ca9bed0e5f8ca396
null
low
[ { "filename": "src/transformers/trainer.py", "patch": "@@ -3056,7 +3056,8 @@ def _load_rng_state(self, checkpoint):\n return\n \n with safe_globals():\n- checkpoint_rng_state = torch.load(rng_file)\n+ check_torch_load_is_safe()\n+ checkpoint_rng_state...
facebook/react
33,447
Enable the `enableEagerAlternateStateNodeCleanup` Feature Flag
## Summary Enables the `enableEagerAlternateStateNodeCleanup` feature flag for all variants, while maintaining the `__VARIANT__` for the internal React Native flavor for backtesting reasons. ## How did you test this change? ``` $ yarn test ```
8b55eb4e724271206bd5dec7dba0a35aedc74493
dddcae7a11b8241cbd6e2de55f9e68881baea458
22
medium
[ { "filename": "packages/shared/forks/ReactFeatureFlags.native-oss.js", "patch": "@@ -48,7 +48,7 @@ export const enableRetryLaneExpiration = false;\n export const enableSchedulingProfiler = __PROFILE__;\n export const enableComponentPerformanceTrack = false;\n export const enableScopeAPI = false;\n-export co...
huggingface/transformers
43,235
Fix typo: seperately → separately in lw_detr converter
## What does this PR do? Fixes a typo in the LW-DETR model conversion script. **Change:** `seperately` → `separately` in a comment **File:** `src/transformers/models/lw_detr/convert_lw_detr_to_hf.py` (line 158) ## Before submitting - [x] This PR fixes a typo or improves the docs (you can dismiss the other checks i...
d8541c36a3fd4c173d8902be382508abf08e0521
c0d2e26fcebfa816db1537773c012f6f9e20a361
6
medium
[ { "filename": "src/transformers/models/lw_detr/convert_lw_detr_to_hf.py", "patch": "@@ -155,7 +155,7 @@\n r\"backbone.0.encoder.blocks.(\\d+).mlp.fc1.(weight|bias)\": r\"backbone.backbone.encoder.layer.\\1.intermediate.fc1.\\2\",\n r\"backbone.0.encoder.blocks.(\\d+).mlp.fc2.(weight|bias)\": r\"...
nodejs/node
61,250
doc: add reusePort error behavior to net module
This PR documents the error behavior for the `reusePort` option in net module. Currently, dgram docs mention that `reusePort` raises an error on unsupported platforms, but net docs don't. This inconsistency can confuse users when their server unexpectedly terminates on platforms like Windows or older Linux. Chang...
null
650f51fe457e2daab3d6be0434b9f93ec7a05275
null
low
[ { "filename": "doc/api/dgram.md", "patch": "@@ -951,7 +951,7 @@ changes:\n port, even if another process has already bound a socket on it. Incoming\n datagrams are distributed to listening sockets. The option is available\n only on some platforms, such as Linux 3.9+, DragonFlyBSD 3.6+, FreeBSD 1...
vuejs/vue
8,586
feat: Store functional render context on vnode in development
Used in https://github.com/vuejs/vue-devtools/pull/719 to display prop values in component inspector.
null
4ecc21c29ec12bb33d3b426cb4d42c579e9b0f2d
null
low
[ { "filename": "src/core/vdom/create-functional-component.js", "patch": "@@ -105,24 +105,27 @@ export function createFunctionalComponent (\n const vnode = options.render.call(null, renderContext._c, renderContext)\n \n if (vnode instanceof VNode) {\n- return cloneAndMarkFunctionalResult(vnode, data, r...
ollama/ollama
9,379
server: validate local path on safetensor create
This PR implements more validation during the safetensor creation process. The changes: - Properly handle relative paths (like `./model.safetensors`) while rejecting absolute paths - Add comprehensive test coverage for various path scenarios including: - Valid relative paths with various notation styles - Nes...
null
bebb6823c03df34404753e42a41e7be8049d3146
null
low
[ { "filename": "server/create.go", "patch": "@@ -8,6 +8,7 @@ import (\n \t\"errors\"\n \t\"fmt\"\n \t\"io\"\n+\t\"io/fs\"\n \t\"log/slog\"\n \t\"net/http\"\n \t\"os\"\n@@ -34,6 +35,7 @@ var (\n \terrOnlyGGUFSupported = errors.New(\"supplied file was not in GGUF format\")\n \terrUnknownType ...
ggml-org/llama.cpp
18,981
opencl: add cumsum op
This PR adds the cumsum op for the OpenCL backend.
null
3d9ab225e71d96458adc2219f0d039bc46dbc8b9
null
low
[ { "filename": "ggml/src/ggml-opencl/CMakeLists.txt", "patch": "@@ -132,6 +132,7 @@ set(GGML_OPENCL_KERNELS\n ssm_conv\n sub\n sum_rows\n+ cumsum\n transpose\n concat\n tsembd", "additions": 1, "deletions": 0 }, { "filename": "ggml/src/ggml-opencl/ggml-opencl.cpp", ...
vuejs/vue
8,709
fix(types): accept `number` type as key on Vue.set/delete (#8707)
<!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-request-guidelines --> <!-- PULL REQUEST TEMPLATE --> <!-- (Update "[ ]" to "[x]" to check a box) --> **What kind of change does this PR introduce?** (check at least one) - [x] Bugf...
c711ec189aaf46399756e34d933ba5e0b6576c36
0ba79e2588309ba386f570ed84d372611c4dd165
2
medium
[ { "filename": "types/test/vue-test.ts", "patch": "@@ -87,8 +87,10 @@ class Test extends Vue {\n this.nextTick(() => {});\n this.nextTick().then(() => {});\n this.set({}, \"\", \"\");\n+ this.set({}, 1, \"\");\n this.set([true, false, true], 1, true);\n this.delete({}, \"\");\n+ thi...
vercel/next.js
88,757
Fix `Build Commands` section in AGENTS.md
The commands must have been hallucinated.
null
e1685de205016c209be30980ac402c43f7cc5f58
null
low
[ { "filename": "AGENTS.md", "patch": "@@ -98,14 +98,14 @@ gt submit --no-edit\n ## Build Commands\n \n ```bash\n-# Build the Next.js package (dev server only - faster)\n-pnpm --filter=next build:dev-server\n+# Build the Next.js package\n+pnpm --filter=next build\n \n # Build everything\n pnpm build\n \n # Ru...
electron/electron
49,482
fix: return early from platform_util::Beep() on Linux if there is no default GDK display
Backport of #49442 See that PR for details. Notes: Fixed crash in platform_util::Beep() on Linux
null
e555e520d3be1e3267e01e51f7eda014bea7c88b
null
low
[ { "filename": "shell/common/platform_util_linux.cc", "patch": "@@ -406,6 +406,8 @@ bool PlatformTrashItem(const base::FilePath& full_path, std::string* error) {\n \n void Beep() {\n auto* display = gdk_display_get_default();\n+ if (!display)\n+ return;\n gdk_display_beep(display);\n }\n ", "addi...
facebook/react
33,446
Do not notify Discord for draft pull requests
When I added the `ready_for_review` event in #32344, no notifications for opened draft PRs were sent due to some other condition. This is not the case anymore, so we need to exclude draft PRs from triggering a notification when the workflow is run because of an `opened` event. This event is still needed because the `re...
null
43714eb4e970d0200fdc5eac887691df7fae53d5
null
low
[ { "filename": ".github/workflows/compiler_discord_notify.yml", "patch": "@@ -11,6 +11,7 @@ permissions: {}\n \n jobs:\n check_access:\n+ if: ${{ github.event.pull_request.draft == false }}\n runs-on: ubuntu-latest\n outputs:\n is_member_or_collaborator: ${{ steps.check_is_member_or_collab...
ollama/ollama
9,431
runner: defer context cancel
defer the cancel to guarantee it runs. e.g. if the http server errors, cancel doesn't currently get called
0c1041ad851d2ce3dc4d74d5fedb1af759956688
31e472baa4d8ffa7396e8f376cbfb60c4410a58e
8
medium
[ { "filename": "runner/llamarunner/runner.go", "patch": "@@ -968,13 +968,14 @@ func Execute(args []string) error {\n \tserver.cond = sync.NewCond(&server.mu)\n \n \tctx, cancel := context.WithCancel(context.Background())\n+\tdefer cancel()\n+\n \tgo server.run(ctx)\n \n \taddr := \"127.0.0.1:\" + strconv.Ito...
huggingface/transformers
43,199
More robust diff checks in `add_dates`
Cc @Cyrilvallez
3377b37ae0caf858f0b576017c2715198c6c5809
3aa89c07f210df18865daee9df81fe2766d13884
19
medium
[ { "filename": "utils/add_dates.py", "patch": "@@ -78,7 +78,13 @@ def check_file_exists_on_github(file_path: str) -> bool:\n def get_modified_cards() -> list[str]:\n \"\"\"Get the list of model names from modified files in docs/source/en/model_doc/\"\"\"\n \n- result = subprocess.check_output([\"git\"...
vuejs/vue
8,848
Remove redundant ternary
**What kind of change does this PR introduce?** - [ ] Bugfix - [ ] Feature - [ ] Code style update - [x] Refactor - [ ] Build-related changes - [ ] Other, please describe: Removed unnecessary ternary operation. **Does this PR introduce a breaking change?** - [ ] Yes - [x] No
2686818beb5728e3b7aa22f47a3b3f0d39d90c8e
374861f72b8b8fda5aa5d54f96bc2f297c775719
6
medium
[ { "filename": "src/core/instance/state.js", "patch": "@@ -224,9 +224,7 @@ export function defineComputed (\n ? createComputedGetter(key)\n : userDef.get\n : noop\n- sharedPropertyDefinition.set = userDef.set\n- ? userDef.set\n- : noop\n+ sharedPropertyDefinition.set = u...
nodejs/node
61,101
src: dump snapshot source with node:generate_default_snapshot_source
This helps diffing snapshots when the reproducibility gets broken. <!-- Before submitting a pull request, please read: - the CONTRIBUTING guide at https://github.com/nodejs/node/blob/HEAD/CONTRIBUTING.md - the commit message formatting guidelines at https://github.com/nodejs/node/blob/HEAD/doc/contributing/p...
null
282d30e526cb0c3ceb7e1720858f61bf6ba7d2cf
null
low
[ { "filename": "src/node.cc", "patch": "@@ -1374,6 +1374,21 @@ ExitCode GenerateAndWriteSnapshotData(const SnapshotData** snapshot_data_ptr,\n DCHECK(snapshot_config.builder_script_path.has_value());\n const std::string& builder_script =\n snapshot_config.builder_script_path.value();\n+\n+ // For ...
ollama/ollama
9,428
fix: replace deprecated functions
these functions are marked as deprecated. replace them with their replacements
a59f66523561dc195a37b78b454d8cd1b8b1fdd7
657685e85d1a187c031ad7722b7395c48d8947b0
18
medium
[ { "filename": "llama/llama.go", "patch": "@@ -262,7 +262,7 @@ func LoadModelFromFile(modelPath string, params ModelParams) (*Model, error) {\n \t\tcparams.progress_callback_user_data = unsafe.Pointer(&handle)\n \t}\n \n-\tm := Model{c: C.llama_load_model_from_file(C.CString(modelPath), cparams)}\n+\tm := Mo...
ggml-org/llama.cpp
20,392
hip: compile debug builds with -O2 on hip to avoid a compiler bug
see the travesty in https://github.com/ROCm/ROCm/issues/5826 for why this is required.
null
d63aa398de455cf8243802c5099dc30a2e26edcc
null
low
[ { "filename": "ggml/src/ggml-hip/CMakeLists.txt", "patch": "@@ -11,6 +11,10 @@ endif()\n list(APPEND CMAKE_PREFIX_PATH ${ROCM_PATH})\n list(APPEND CMAKE_PREFIX_PATH \"${ROCM_PATH}/lib64/cmake\")\n \n+if (NOT DEFINED CMAKE_HIP_FLAGS_DEBUG)\n+ set(CMAKE_HIP_FLAGS_DEBUG \"-g -O2\")\n+endif()\n+\n # CMake o...
vercel/next.js
88,749
[test] Deflake `test/development/app-dir/hydration-error-count/hydration-error-count.test.ts`
`toDisplayCollapsedRedbox` was just used to avoid having to update the snapshot with internals when we were burning down Cache Component issues. With https://github.com/vercel/next.js/issues/85420, we are coupled less to Next.js implementation detail and can unfork the assertion. Fixes https://github.com/vercel/next...
null
97532172c065ca2b49dd79d1fa44c3a0b864065b
null
low
[ { "filename": "test/development/app-dir/hydration-error-count/hydration-error-count.test.ts", "patch": "@@ -334,11 +334,11 @@ describe('hydration-error-count', () => {\n it('should display runtime error separately from hydration errors', async () => {\n const browser = await next.browser('/hydration-w...
facebook/react
33,443
[Flight] Add a cached 3rd-party component to the Flight fixture
This should allow us to visualize what https://github.com/facebook/react/pull/33438 is trying to convey. An uncached 3rd-party component is displayed like this in the dev tools: <img width="1072" alt="Screenshot 2025-06-05 at 12 57 32" src="https://github.com/user-attachments/assets/d418ae23-d113-4dc9-98b8-ab4267...
5717f1933f2e8b10406fde1043c3047cbfbddc82
b4477d3800ccb0bdf26670cd1f021d094159c38f
21
medium
[ { "filename": "fixtures/flight/src/App.js", "patch": "@@ -1,4 +1,7 @@\n import * as React from 'react';\n+import {renderToPipeableStream} from 'react-server-dom-webpack/server';\n+import {createFromNodeStream} from 'react-server-dom-webpack/client';\n+import {PassThrough, Readable} from 'stream';\n \n impor...
electron/electron
49,489
fix: potential devtools crash on empty
Backport of #49468 See that PR for details. Notes: none
null
6ce52ad79290297495442c731d7d7b3f579da44d
null
low
[ { "filename": "lib/renderer/inspector.ts", "patch": "@@ -11,12 +11,14 @@ const { contextIsolationEnabled } = internalContextBridge;\n * 1) Use menu API to show context menu.\n */\n window.onload = function () {\n- if (contextIsolationEnabled) {\n- internalContextBridge.overrideGlobalValueFromIsolatedWor...
vuejs/vue
9,119
fix #9114 correct the compiled code for multiline v-if in scoped slots
add wrapping paranthesis for v-if conditions the generated code for elements with template in scopedslots was wrong when multiline v-if starting with a newline was used fix #9114 <!-- Please make sure to read the Pull Request Guidelines: https://github.com/vuejs/vue/blob/dev/.github/CONTRIBUTING.md#pull-req...
null
ef8524ab7db8d64ac449ce74f5858aa9d91357ad
null
low
[ { "filename": "src/compiler/codegen/index.js", "patch": "@@ -367,7 +367,7 @@ function genScopedSlot (\n const fn = `function(${String(el.slotScope)}){` +\n `return ${el.tag === 'template'\n ? el.if\n- ? `${el.if}?${genChildren(el, state) || 'undefined'}:undefined`\n+ ? `(${el.if})?...
ollama/ollama
9,426
cmake: build compute capability 12.0
Focusing initial Blackwell support on 12.0 which includes the 50x series of GeForce cards. In the future additional compute capabilities may be added.
e12af460eda05120f78d8066dd06d9ec1e33055a
a14912858e71392e9240c87e7e162d5d203dcb4e
23
medium
[ { "filename": "CMakePresets.json", "patch": "@@ -28,7 +28,7 @@\n \"name\": \"CUDA 12\",\n \"inherits\": [ \"CUDA\" ],\n \"cacheVariables\": {\n- \"CMAKE_CUDA_ARCHITECTURES\": \"50;60;61;70;75;80;86;87;89;90;90a;100\"\n+ \"CMAKE_CUDA_ARCHITECTURES\": \"50;60;61;70;75;80;86;87;...
huggingface/transformers
43,182
Fix failing `Owlv2ModelIntegrationTest` & `OwlViTModelIntegrationTest`
# What does this PR do? Fixes failing [Owlv2ModelIntegrationTest](https://github.com/huggingface/transformers/actions/runs/20706072009/job/59437072091#step:14:4172) & [OwlViTModelIntegrationTest](https://github.com/huggingface/transformers/actions/runs/20706072009/job/59437072113#step:14:4015) <img width="1351" h...
null
61d7f8a4495822604f14b621beb65f5e41bebee5
null
low
[ { "filename": "tests/models/owlv2/test_modeling_owlv2.py", "patch": "@@ -735,7 +735,7 @@ def test_inference_interpolate_pos_encoding(self):\n [-3.3644, -4.0717, -4.0717, -4.0717],\n [-2.9425, -4.0717, -4.0717, -4.0717],\n ]\n- )\n+ ).to(torch_device)...
facebook/react
33,403
[Flight] Encode Async I/O Tasks using the Enclosing Line/Column
Stacked on #33402. There's a bug in Chrome Performance tracking which uses the enclosing line/column instead of the callsite in stacks. For our fake eval:ed functions that represents functions on the server, we can position the enclosing function body at the position of the callsite to simulate getting the right ...
9cc74fec749bcca2e0f5d1e41aa612b2135641ff
154008172573d64519ebbc23da611a27073b0a8c
1
medium
[ { "filename": "packages/react-client/src/ReactFlightClient.js", "patch": "@@ -813,7 +813,13 @@ function createElement(\n console,\n getTaskName(type),\n );\n- const callStack = buildFakeCallStack(response, stack, env, createTaskFn);\n+ const callStack = buildFakeCallStack(\n+...