Robert Smith PRO
tcclaviger
AI & ML interests
Local SWE supporting LLMs that can be used for private data. Currently expanding capabilities to enable use of larger models supported by a custom tool set environment providing context 7 like knowledge with added DSA, Software design, and web crawling capabilities to enable autonomous expansion of knowledge as the LLM deems necessary.
Recent Activity
new activity about 18 hours ago
stepfun-ai/Step-3.7-Flash:How to turn off thinking.... new activity about 18 hours ago
tcclaviger/Step-3.7-Flash-240REAP:How to make thinking stop, so you can use none, low, mid, high thinking new activity about 19 hours ago
DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking:Is this model reliable for serious work?Organizations
None yet
How to turn off thinking....
2
#14 opened about 18 hours ago
by tcclaviger
How to make thinking stop, so you can use none, low, mid, high thinking
#1 opened about 18 hours ago
by tcclaviger
New activity in DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking about 19 hours ago
Is this model reliable for serious work?
3
#7 opened about 21 hours ago
by 
Nunodonato
Supers Found in model
7
#11 opened 3 days ago
by tcclaviger
MTP should be left in place, its trivial to add back in and it is beneficial.
👍👀 3
12
#4 opened about 1 month ago
by tcclaviger
New activity in DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF 21 days ago
Qwen3.6-27B Local Deployment: llama-server Parameter Tuning Consultation
3
#8 opened 23 days ago
by 
jason888888
Bravo
❤️ 2
2
#1 opened 23 days ago
by tcclaviger
MTP Grafted back in, findings...
4
#6 opened about 1 month ago
by tcclaviger
Is MTP possible?
3
#2 opened about 1 month ago
by 
JDWarner
About this FP8
1
#1 opened about 1 month ago
by 
sudo-0x2a
Model loading reduction from 5.5 minutes to 5.26 seconds
1
#19 opened 2 months ago
by tcclaviger
40B to 80B dense LLM: A better/smarter model than 122B that can still fit
1
#16 opened 3 months ago
by 
rtzurtz
Love the quant! Nicely done.
❤️ 1
3
#2 opened 2 months ago
by tcclaviger
MTP Working on AMD 4xR9700 and MXFP4 Custom Kernel
👍 1
1
#18 opened 2 months ago
by tcclaviger
mlx-community/MiniMax-M2.1-4bit templates issue
2
#12 opened 5 months ago
by 
marutichintan
Official MXFP4 Quant please
👀🔥 16
1
#8 opened 6 months ago
by tcclaviger
Add GPTQModel W4A16 INT format Quant ?
1
#13 opened 9 months ago
by 
X-SZM
New activity in tcclaviger/Qwen3-Coder-42B-A3B-Instruct-TOTAL-RECALL-MASTER-CODER-M-512k-ctx-W4A16 7 months ago
Currently in production on private service.
#1 opened 7 months ago
by tcclaviger
Running on 2xAI Pro R9700s via ROCM 7 and nightly VLLM with this
🤝 1
8
#1 opened 8 months ago
by tcclaviger