Reasoning models you can run locally (62 models)
← All models

Collection

Reasoning models

62 catalogued models. Coverage shows how far LocalAIReady can answer compatibility for each one.

Model Publisher Parameters Family Coverage
LFM2 350M Math Liquid AI 354M Lfm Partial
Falcon-H1-Tiny R 0.6B Technology Innovation Institute 622M Falcon Partial
LFM2.5 1.2B Thinking Liquid AI 1.2B Lfm Partial
Qwen2-Math 1.5B Instruct Alibaba Qwen 1.5B Qwen Full
DictaLM 3.0 1.7B Thinking Dicta 1.7B Dictalm Full
Palmyra Mini Thinking A Writer 1.8B Palmyra Full
DeepScaleR 1.5B Preview Agentica 1.8B Agentica Full
DeepSeek-R1-Distill-Qwen-1.5B DeepSeek 1.8B Deepseek Full
Palmyra Mini Thinking B Writer 1.8B Palmyra Full
EXAONE Deep 2.4B LG AI Research 2.4B Exaone Full
MediX R1 2B MBZUAI 2.4B Medix Full
Jamba Reasoning 3B AI21 Labs 3.2B Jamba Partial
llama3.2 typhoon2 t1 3B research preview Typhoon (SCB 10X) 3.2B Typhoon Full
SmallThinker 3B Preview PowerInfer 3.4B Smallthinker Full
Phi-4-mini-reasoning Microsoft 3.8B Phi Full
Qwen3 4B Thinking 2507 Alibaba Qwen 4B Qwen Full
Ministral 3 3B Reasoning 2512 Mistral AI 4.3B Mistral Full
typhoon si med thinking 4B research preview Typhoon (SCB 10X) 4.4B Typhoon Full
Mathstral 7B v0.1 Mistral AI 7.2B Mistral Full
Falcon-H1R 7B Technology Innovation Institute 7.6B Falcon Partial
DeepSeek-R1-Distill-Qwen-7B DeepSeek 7.6B Deepseek Full
Marco-o1 Alibaba AIDC-AI 7.6B Marco Full
Qwen2-Math 7B Instruct Alibaba Qwen 7.6B Qwen Full
EXAONE Deep 7.8B LG AI Research 7.8B Exaone Full
DeepSeek-R1-Distill-Llama-8B DeepSeek 8B Deepseek Full
Llama-SEA-LION v3.5 8B R AI Singapore 8B Sea-lion Full
Foundation Sec 8B Reasoning Foundation AI (Cisco) 8B Foundation-sec Full
MiniCPM4.1 8B OpenBMB 8.2B Minicpm Full
DeepSeek-R1-0528-Qwen3-8B DeepSeek 8.2B Deepseek Full
llm jp 4 8B thinking LLM-jp 8.6B Llm-jp Full
MediX R1 8B MBZUAI 8.8B Medix Full
Ministral 3 8B Reasoning 2512 Mistral AI 8.9B Mistral Full
Turkish Gemma 9B T1 YTU COSMOS 9.2B Turkish-llm Full
Homunculus Arcee AI 12.5B Arcee Full
Ministral 3 14B Reasoning 2512 Mistral AI 13.9B Mistral Full
Phi-4-reasoning Microsoft 14.7B Phi Full
Phi-4-reasoning-plus Microsoft 14.7B Phi Full
DeepSeek-R1-Distill-Qwen-14B DeepSeek 14.8B Deepseek Full
Apriel 1.6 15B Thinker ServiceNow 14.9B Apriel Full
Ring mini 2.0 inclusionAI (Ant Group) 16.3B Ling Partial
DictaLM 3.0 24B Thinking Dicta 23.6B Dictalm Full
Magistral Small 2506 Mistral AI 23.6B Mistral Full
Magistral Small 2507 Mistral AI 23.6B Mistral Full
Magistral Small 2509 Mistral AI 24B Mistral Full
Qwen3 30B-A3B Thinking 2507 Alibaba Qwen 30.5B Qwen Full
MediX R1 30B MBZUAI 31.1B Medix Full
EXAONE Deep 32B LG AI Research 32B Exaone Full
EXAONE 4.0 32B LG AI Research 32B Exaone Full
llm jp 4 32B a3b thinking LLM-jp 32.1B Llm-jp Full
DeepSeek-R1-Distill-Qwen-32B DeepSeek 32.8B Deepseek Full
OpenThinker2 32B Open Thoughts 32.8B Openthinker Full
QwQ-32B Alibaba Qwen 32.8B Qwen Full
QwQ 32B Preview Alibaba Qwen 32.8B Qwen Full
deepseek r1 distill qwen2.5 bakeneko 32B rinna 32.8B Rinna Full
llm jp 4 33B thinking LLM-jp 33.2B Llm-jp Full
Hermes 4.3 36B Nous Research 36.2B Hermes Full
DeepSeek-R1-Distill-Llama-70B DeepSeek 70.6B Deepseek Full
Llama-SEA-LION v3.5 70B R AI Singapore 70.6B Sea-lion Full
Qwen2-Math 72B Instruct Alibaba Qwen 72.7B Qwen Full
Qwen3-Next 80B-A3B Thinking Alibaba Qwen 81.3B Qwen Partial
Qwen3 235B-A22B Thinking 2507 Alibaba Qwen 235B Qwen Full
DeepSeek-R1-0528 DeepSeek 685B Deepseek Full

What else can your PC run?

Pick your GPU and system RAM to see every catalogued model that fits, with tested, estimated and potential answers kept apart.

Check my PC