Instructions to use continuum-ai/olmoe-1b-7b-compacted-5b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use continuum-ai/olmoe-1b-7b-compacted-5b with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("continuum-ai/olmoe-1b-7b-compacted-5b")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

llama-cpp-python

How to use continuum-ai/olmoe-1b-7b-compacted-5b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="continuum-ai/olmoe-1b-7b-compacted-5b",
	filename="olmoe-1b-7b-compacted-5b.Q5_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use continuum-ai/olmoe-1b-7b-compacted-5b with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M
# Run inference directly in the terminal:
llama-cli -hf continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M
# Run inference directly in the terminal:
llama-cli -hf continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M
# Run inference directly in the terminal:
./llama-cli -hf continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M

Use Docker

docker model run hf.co/continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M

LM Studio
Jan

vLLM

How to use continuum-ai/olmoe-1b-7b-compacted-5b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "continuum-ai/olmoe-1b-7b-compacted-5b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "continuum-ai/olmoe-1b-7b-compacted-5b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M

Ollama
How to use continuum-ai/olmoe-1b-7b-compacted-5b with Ollama:
```
ollama run hf.co/continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M
```

Unsloth Studio new

How to use continuum-ai/olmoe-1b-7b-compacted-5b with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for continuum-ai/olmoe-1b-7b-compacted-5b to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for continuum-ai/olmoe-1b-7b-compacted-5b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for continuum-ai/olmoe-1b-7b-compacted-5b to start chatting

MLX LM

How to use continuum-ai/olmoe-1b-7b-compacted-5b with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "continuum-ai/olmoe-1b-7b-compacted-5b"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "continuum-ai/olmoe-1b-7b-compacted-5b"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "continuum-ai/olmoe-1b-7b-compacted-5b",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use continuum-ai/olmoe-1b-7b-compacted-5b with Docker Model Runner:
```
docker model run hf.co/continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M
```

Lemonade

How to use continuum-ai/olmoe-1b-7b-compacted-5b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull continuum-ai/olmoe-1b-7b-compacted-5b:Q5_K_M

Run and chat with the model

lemonade run user.olmoe-1b-7b-compacted-5b-Q5_K_M

List all available models

lemonade list

25% Experts Pruned, 36.0 HUMANEVAL (base 40.9)

OLMoE-1B-7B-0924-Instruct compacted via per-layer-normalized MoE expert pruning against the unmodified teacher.

HUMANEVAL: 36.0 (base 40.9, Δ -4.9)
HUMANEVAL+PLUS: 31.7 (base 36.6, Δ -4.9)

Every claim on this card is verified
Trust: self-attested · 2 benchmarks · 1 device tested
ForgeAlloy chain of custody · Download alloy · Merkle-chained

Cross-architecture validation artifact for §4.1.3.4. OLMoE-1B-7B (the smallest serious MoE on HF, fully-open Allen AI release) compacted from 7B to ~5B via calibration-aware MoE expert pruning on a held-out Python code corpus. Hardware-measured 36.0 HumanEval against unmodified base 40.9 (Δ −4.9, both Q5_K_M on the same 5090). The forge methodology that produced qwen3-coder-30b-a3b-compacted-19b-256k ports to a structurally distinct MoE family (OlmoeForCausalLM vs Qwen3MoeForCausalLM) without any modification to the forge scripts. The negative-baseline broad-corpus variant scored 28.0 — the +8.0 swing from changing only the calibration corpus is the lever §4.1.3.4 names. This is a methodology proof point, not a tier-leading artifact; OLMoE is general-purpose, not coder-specific, so HumanEval is not its strength. Use the qwen3-coder-30b-a3b artifact if you need a fits-12-GB code model.

Benchmarks

Benchmark	Score	Base	Δ	Verified
humaneval	36.0	40.9	-4.9	✅ Result hash
humaneval_plus	31.7	36.6	-4.9	✅ Result hash

What Changed (Base → Forged)

	Base	Forged	Delta
Pipeline		expert-activation-profile → expert-prune → quant → eval	1 cycles

Runs On

Device	Format	Size	Speed
NVIDIA GeForce RTX 5090	Q5_K_M	3.6GB	Verified
MacBook Pro 32GB	fp16	3.6GB	Expected
MacBook Air 16GB	Q8_0	~1.8GB	Expected
MacBook Air 8GB	Q4_K_M	~1.1GB	Expected
iPhone / Android	Q4_K_M	~1.1GB	Expected

Quick Start

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("continuum-ai/olmoe-1b-7b-compacted-5b",
    torch_dtype="auto", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("continuum-ai/olmoe-1b-7b-compacted-5b")

inputs = tokenizer("def merge_sort(arr):", return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Methodology

Produced via MoE expert pruning, GGUF quantization. Full methodology, ablations, and per-stage rationale are in the methodology paper and the companion MODEL_METHODOLOGY.md in this repository. The pipeline ran as expert-activation-profile → expert-prune → quant → eval over 1 cycle on NVIDIA GeForce RTX 5090.

Limitations

HumanEval is not OLMoE's natural benchmark. OLMoE is general-purpose (Allen AI), not coder-specific. The 40.9 base / 36.0 student numbers are methodology validation, not tier-leading absolute quality. For a tier-leading code model, see qwen3-coder-30b-a3b-compacted-19b-256k.
Validates §4.1.3.4 cross-architecture; does NOT compete on absolute numbers. This is the second empirical anchor for the methodology paper, alongside the Qwen3-Coder-30B-A3B v1. Together they demonstrate that the activation-count importance metric is architecture-invariant across two structurally distinct MoE families.
Calibration corpus was 300 Python code examples. For non-code workloads (math/reasoning/general), the methodology will preserve OLMoE's general capability if profiled on a matching corpus — but that's a separate forge run.
Single GGUF tier shipped (Q5_K_M, 3.6 GB). Q4_K_M and Q8_0 will be added in v1.1 if there's demand.

Chain of Custody

Scan the QR or verify online. Download the alloy file to verify independently.

What	Proof
Model weights	`sha256:7f3b3c31279035cd5226f13cd602875ba...`
Forged on	NVIDIA GeForce RTX 5090, ?
Published	huggingface — 2026-04-08T16:36:55.037319+00:00
Trust level	`self-attested`
Spec	ForgeAlloy — Rust/Python/TypeScript

Make Your Own

Forged with Continuum — a distributed AI world that runs on your hardware.

The Factory configurator lets you design and forge custom models visually — context extension, pruning, LoRA, quantization, vision/audio modalities. Pick your target devices, the system figures out what fits.

GitHub · All Models · Forge-Alloy