Comparison of Open Source Models
Comparison and analysis of open source AI models across key performance metrics including quality, performance, inference speed, context window, parameter count & licensing details. Models are considered Open Source (also commonly referred to as open weights) where their weights are accessible to download. This allows self-hosting on your own infrastructure and enables customizing the model such as through fine-tuning. Click on any model to see detailed metrics. For more details relating to our methodology, see our FAQs.
gpt-oss-120B (high) and 
DeepSeek V3.1 Terminus are the highest intelligence open source models, followed by 
Qwen3 235B A22B 2507 & DeepSeek V3.2 Exp.
Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Estimate (independent evaluation forthcoming)
Total Parameters
Trainable parameters in billions
Further details
| Weights | Provider Benchmarks | |||||||
|---|---|---|---|---|---|---|---|---|
gpt-oss-120B (high) OpenAI | 58 | 117B (5.1B active at inference time) | 131k | $0.3 | 309 | 🤗 |    +17 more | View |
DeepSeek V3.1 Terminus (Reasoning) DeepSeek | 58 | 685B (37B active at inference time) | 128k | $1.9 | - | 🤗 |   | View |
Qwen3 235B A22B 2507 (Reasoning) Alibaba | 57 | 235B (22B active at inference time) | 256k | $2.6 | 84 | 🤗 |    +6 more | View |
DeepSeek V3.2 Exp (Reasoning) DeepSeek | 57 | 685B (37.0B active at inference time) | 128k | $0.3 | 26 | 🤗 | | View |
 GLM-4.6 (Reasoning) Z AI | 56 | 357B (32B active at inference time) | 200k | $1.0 | 51 | 🤗 |   +1 more | View |
Qwen3 Next 80B A3B (Reasoning) Alibaba | 54 | 80B (3B active at inference time) | 262k | $1.9 | 184 | 🤗 |  +4 more | View |
DeepSeek V3.1 (Reasoning) DeepSeek | 54 | 685B (37.0B active at inference time) | 128k | $0.7 | - | 🤗 | +1 more | View |
DeepSeek R1 0528 (May '25) DeepSeek | 52 | 685B (37B active at inference time) | 128k | $1.6 | - | 🤗 |   +12 more | View |
 Apriel-v1.5-15B-Thinker ServiceNow | 52 | 15B | 128k | - | - | Not available | - | View |
GLM-4.5 (Reasoning) Z AI | 51 | 355B (32B active at inference time) | 128k | $1.0 | 76 | 🤗 |  +4 more | View |
 Kimi K2 0905 Moonshot AI | 50 | 1.0KB (32.0B active at inference time) | 256k | $1.3 | 55 | 🤗 | +3 more | View |
GLM-4.5-Air Z AI | 49 | - | 128k | $0.4 | 211 | 🤗 | +2 more | View |
Kimi K2 Moonshot AI | 48 | 1.0KB (32B active at inference time) | 128k | $1.1 | 55 | 🤗 | +6 more | View |
Qwen3 30B A3B 2507 (Reasoning) Alibaba | 46 | 30.5B (3.3B active at inference time) | 262k | $0.8 | 101 | 🤗 | | View |
DeepSeek V3.2 Exp (Non-reasoning) DeepSeek | 46 | 685B (37.0B active at inference time) | 128k | $0.3 | 24 | 🤗 | +1 more | View |
 MiniMax M1 80k MiniMax | 46 | 456B (45.9B active at inference time) | 1.00M | $0.8 | - | 🤗 | | View |
DeepSeek V3.1 Terminus (Non-reasoning) DeepSeek | 46 | 685B (37B active at inference time) | 128k | $0.5 | - | 🤗 | +2 more | View |
Qwen3 235B A22B 2507 Instruct Alibaba | 45 | 235B (22.0B active at inference time) | 256k | $1.2 | 47 | 🤗 |  +9 more | View |
 Llama Nemotron Super 49B v1.5 (Reasoning) NVIDIA | 45 | 49B | 128k | $0.2 | 80 | 🤗 | ? | View |
Qwen3 Next 80B A3B Instruct Alibaba | 45 | 80B (3B active at inference time) | 262k | $0.9 | 161 | 🤗 | +3 more | View |
DeepSeek V3.1 (Non-reasoning) DeepSeek | 45 | 685B (37B active at inference time) | 128k | $0.9 | - | 🤗 |  +6 more | View |
GLM-4.6 (Non-reasoning) Z AI | 45 | 357B (32B active at inference time) | 200k | $1.0 | 51 | 🤗 | | View |
DeepSeek R1 (Jan '25) DeepSeek | 44 | 685B (37B active at inference time) | 128k | $2.0 | - | 🤗 | +8 more | View |
Qwen3 4B 2507 (Reasoning) Alibaba | 43 | 4.02B | 262k | - | - | 🤗 | - | View |
gpt-oss-20B (high) OpenAI | 43 | 21B (3.6B active at inference time) | 131k | $0.1 | 156 | 🤗 | +9 more | View |
 Magistral Small 1.2 Mistral | 43 | 24B | 128k | $0.8 | 218 | 🤗 | | View |
 EXAONE 4.0 32B (Reasoning) LG AI Research | 43 | 32B | 131k | $0.7 | 67 | 🤗 |  | View |
Qwen3 Coder 480B A35B Instruct Alibaba | 42 | 480B (35.0B active at inference time) | 262k | $3.0 | 50 | 🤗 |  +10 more | View |
Qwen3 235B A22B (Reasoning) Alibaba | 42 | 235B (22B active at inference time) | 32.8k | $2.6 | 57 | 🤗 | +2 more | View |
MiniMax M1 40k MiniMax | 42 | 456B (45.9B active at inference time) | 1.00M | $0.8 | - | 🤗 | | View |
 Hermes 4 - Llama-3.1 405B (Reasoning) Nous Research | 42 | 406B | 128k | $1.5 | 37 | 🤗 | | View |
DeepSeek V3 0324 DeepSeek | 41 | 671B (37B active at inference time) | 128k | $0.9 | - | 🤗 | +10 more | View |
Qwen3 Omni 30B A3B (Reasoning) Alibaba | 40 | 35.3B (3.0B active at inference time) | 65.5k | $0.4 | 88 | 🤗 | | View |
Hermes 4 - Llama-3.1 70B (Reasoning) Nous Research | 39 | 70.6B | 128k | $0.2 | 85 | 🤗 | | View |
Qwen3 32B (Reasoning) Alibaba | 39 | 32.8B | 32.8k | $2.6 | 57 | 🤗 | +5 more | View |
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) NVIDIA | 38 | 253B | 128k | $0.9 | 38 | 🤗 | | View |
QwQ 32B Alibaba | 38 | 32.8B | 131k | $0.5 | 56 | 🤗 | +3 more | View |
GLM-4.5V (Reasoning) Z AI | 37 | 108B (12.0B active at inference time) | 64.0k | $0.9 | - | 🤗 | | View |
Qwen3 30B A3B 2507 Instruct Alibaba | 37 | 30.5B (3.3B active at inference time) | 262k | $0.3 | 86 | 🤗 | | View |
NVIDIA Nemotron Nano 9B V2 (Reasoning) NVIDIA | 37 | 9B | 131k | $0.1 | 125 | 🤗 | | View |
Qwen3 30B A3B (Reasoning) Alibaba | 37 | 30.5B (3.3B active at inference time) | 32.8k | $0.8 | 71 | 🤗 | +3 more | View |
NVIDIA Nemotron Nano 9B V2 (Non-reasoning) NVIDIA | 36 | 9B | 131k | $0.1 | 125 | 🤗 | | View |
Qwen3 14B (Reasoning) Alibaba | 36 | 14.8B | 32.8k | $1.3 | 54 | 🤗 | | View |
 Llama 4 Maverick Meta | 36 | 402B (17B active at inference time) | 1.00M | $0.4 | 142 | 🤗 | +12 more | View |
Llama 3.3 Nemotron Super 49B v1 (Reasoning) NVIDIA | 35 | 49B | 128k | - | - | 🤗 | - | View |
DeepSeek R1 0528 Qwen3 8B DeepSeek | 35 | 8.19B | 32.8k | $0.1 | 39 | 🤗 | | View |
Qwen3 Coder 30B A3B Instruct Alibaba | 33 | 30.5B (3.3B active at inference time) | 262k | $0.9 | 95 | 🤗 | +1 more | View |
DeepSeek R1 Distill Qwen 32B DeepSeek | 33 | 32B | 128k | $0.3 | 31 | 🤗 | | View |
Hermes 4 - Llama-3.1 405B (Non-reasoning) Nous Research | 33 | 406B | 128k | $1.5 | 34 | 🤗 | | View |
 Reka Flash 3 Reka AI | 33 | 21B | 128k | $0.3 | 51 | 🤗 | | View |
DeepSeek V3 (Dec '24) DeepSeek | 32 | 671B (37B active at inference time) | 128k | $0.8 | - | 🤗 | +4 more | View |
Magistral Small 1 Mistral | 32 | 23.6B | 40.0k | $0.8 | 208 | 🤗 | | View |
DeepSeek R1 Distill Llama 70B DeepSeek | 31 | 70B | 128k | $0.8 | 121 | 🤗 | +3 more | View |
Qwen3 4B 2507 Instruct Alibaba | 30 | 4.02B | 262k | - | - | 🤗 | - | View |
EXAONE 4.0 32B (Non-reasoning) LG AI Research | 30 | 32B | 131k | $0.7 | 58 | 🤗 | | View |
Qwen3 Omni 30B A3B Instruct Alibaba | 30 | 35.3B (3.0B active at inference time) | 65.5k | $0.4 | 80 | 🤗 | | View |
Qwen3 235B A22B (Non-reasoning) Alibaba | 30 | 235B (22B active at inference time) | 32.8k | $1.2 | 47 | 🤗 | +2 more | View |
DeepSeek R1 Distill Qwen 14B DeepSeek | 30 | 14B | 128k | $0.9 | 105 | 🤗 | | View |
Qwen3 14B (Non-reasoning) Alibaba | 29 | 14.8B | 32.8k | $0.6 | 51 | 🤗 | | View |
Mistral Small 3.2 Mistral | 29 | 24B | 128k | $0.1 | 103 | 🤗 | | View |
Qwen2.5 Instruct 72B Alibaba | 29 | 72B | 131k | - | 49 | 🤗 | +2 more | View |
Qwen3 8B (Reasoning) Alibaba | 28 | 8.19B | 131k | $0.7 | 88 | 🤗 | | View |
Llama 4 Scout Meta | 28 | 109B (17B active at inference time) | 10.0M | $0.3 | 112 | 🤗 | +10 more | View |
 Command A Cohere | 28 | 111B | 256k | $4.4 | 94 | 🤗 | | View |
QwQ 32B-Preview Alibaba | 28 | 32.8B | 32.8k | $0.7 | 47 | 🤗 | | View |
Llama 3.3 Instruct 70B Meta | 28 | 70B | 128k | $0.6 | 89 | 🤗 | +19 more | View |
Devstral Small (Jul '25) Mistral | 27 | 24B | 256k | $0.1 | 158 | 🤗 | | View |
Mistral Large 2 (Nov '24) Mistral | 27 | 123B | 128k | $3.0 | 46 | 🤗 | | View |
Exaone 4.0 1.2B (Reasoning) LG AI Research | 27 | 1.28B | 64.0k | - | - | 🤗 | - | View |
Llama Nemotron Super 49B v1.5 (Non-reasoning) NVIDIA | 27 | 49B | 128k | $0.2 | 74 | 🤗 | ? | View |
Qwen3 30B A3B (Non-reasoning) Alibaba | 26 | 30.5B (3.3B active at inference time) | 32.8k | $0.3 | 63 | 🤗 | | View |
Qwen3 32B (Non-reasoning) Alibaba | 26 | 32.8B | 32.8k | $1.2 | 52 | 🤗 | +7 more | View |
Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning) NVIDIA | 26 | 4.51B | 128k | - | - | 🤗 | - | View |
GLM-4.5V (Non-reasoning) Z AI | 26 | 108B (12.0B active at inference time) | 64.0k | $0.9 | 73 | 🤗 | | View |
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) NVIDIA | 26 | 49B | 128k | - | - | 🤗 | - | View |
MiniMax-Text-01 MiniMax | 26 | 456B (45.9B active at inference time) | 4.00M | $0.4 | - | 🤗 | | View |
Llama 3.1 Instruct 405B Meta | 26 | 405B | 128k | $3.5 | 35 | 🤗 | ? +8 more | View |
Qwen3 4B (Reasoning) Alibaba | 26 | 4.02B | 32.0k | $0.4 | 92 | 🤗 | | View |
 Llama 3.1 Tulu3 405B Allen Institute for AI | 25 | 405B | 128k | - | - | 🤗 | - | View |
Pixtral Large Mistral | 25 | 124B | 128k | $3.0 | 46 | 🤗 | | View |
 Grok 2 (Dec '24) xAI | 25 | 270B | 131k | - | - | 🤗 | - | View |
Phi-4 Microsoft Azure | 25 | 14B | 16.0k | $0.2 | 34 | 🤗 | | View |
Hermes 4 - Llama-3.1 70B (Non-reasoning) Nous Research | 24 | 70.6B | 128k | $0.2 | 78 | 🤗 | | View |
Llama 3.1 Nemotron Instruct 70B NVIDIA | 24 | 70B | 128k | $0.6 | 33 | 🤗 | | View |
Mistral Small 3.1 Mistral | 23 | 24B | 128k | $0.1 | 170 | 🤗 | +2 more | View |
Qwen3 8B (Non-reasoning) Alibaba | 23 | 8.19B | 32.8k | $0.3 | 81 | 🤗 | | View |
Qwen2.5 Instruct 32B Alibaba | 23 | 32B | 128k | - | - | 🤗 | - | View |
 Granite 4.0 H Small IBM | 23 | 32B (9B active at inference time) | 128k | $0.1 | 51 | Not available |  | View |
Llama 3.1 Instruct 70B Meta | 23 | 70B | 128k | $0.8 | 102 | 🤗 | ? +8 more | View |
Qwen3 1.7B (Reasoning) Alibaba | 22 | 2.03B | 32.0k | $0.4 | 123 | 🤗 | | View |
Mistral Large 2 (Jul '24) Mistral | 22 | 123B | 128k | $3.0 | 33 | 🤗 | | View |
Qwen2.5 Coder Instruct 32B Alibaba | 22 | 32B | 131k | $0.2 | 55 | 🤗 | | View |
Gemma 3 27B Instruct Google | 22 | 27.4B | 128k | - | 52 | 🤗 | | View |
Mistral Small 3 Mistral | 21 | 24B | 32.0k | $0.1 | 204 | 🤗 | | View |
 Jamba 1.7 Large AI21 Labs | 21 | 398B (94.0B active at inference time) | 256k | $3.5 | 44 | 🤗 | | View |
DeepSeek-V2.5 (Dec '24) DeepSeek | 21 | 236B (21B active at inference time) | 128k | - | - | 🤗 | - | View |
Qwen3 4B (Non-reasoning) Alibaba | 21 | 4.02B | 32.0k | $0.2 | 87 | 🤗 | | View |
Exaone 4.0 1.2B (Non-reasoning) LG AI Research | 20 | 1.28B | 64.0k | - | - | 🤗 | - | View |
Gemma 3 12B Instruct Google | 20 | 12.2B | 128k | - | 52 | 🤗 | +1 more | View |
DeepSeek-V2.5 DeepSeek | 20 | 236B (21B active at inference time) | 128k | - | - | 🤗 | - | View |
Devstral Small (May '25) Mistral | 20 | 23.6B | 256k | $0.1 | 106 | 🤗 | | View |
DeepSeek R1 Distill Llama 8B DeepSeek | 19 | 8B | 128k | - | - | 🤗 | - | View |
 R1 1776 Perplexity | 19 | 671B (37B active at inference time) | 128k | $3.5 | - | 🤗 | | View |
Llama 3.2 Instruct 90B (Vision) Meta | 19 | 90B | 128k | $0.7 | 37 | 🤗 | +1 more | View |
 Solar Mini Upstage | 19 | 10.7B | 4.10k | $0.1 | 81 | 🤗 | | View |
Grok-1 xAI | 18 | 314B (78B active at inference time) | 8.19k | - | - | 🤗 | - | View |
Qwen2 Instruct 72B Alibaba | 18 | 72B | 131k | - | - | 🤗 | - | View |
Gemma 2 27B Google | 17 | 27.2B | 8.19k | - | - | 🤗 | - | View |
Llama 3.1 Instruct 8B Meta | 17 | 8B | 128k | $0.1 | 179 | 🤗 |  +18 more | View |
Granite 4.0 Micro IBM | 16 | 3B | 128k | - | - | Not available | - | View |
Phi-4 Mini Instruct Microsoft Azure | 16 | 3.84B | 128k | - | 46 | 🤗 | | View |
DeepHermes 3 - Mistral 24B Preview (Non-reasoning) Nous Research | 16 | 24B | 32.0k | - | - | 🤗 | - | View |
Llama 3.2 Instruct 11B (Vision) Meta | 16 | 11B | 128k | $0.2 | 71 | 🤗 | | View |
Gemma 3n E4B Instruct Google | 15 | 8.39B (4.0B active at inference time) | 32.0k | $0.0 | 50 | 🤗 | | View |
Granite 3.3 8B (Non-reasoning) IBM | 15 | 8.17B | 128k | $0.1 | 510 | 🤗 | | View |
Jamba 1.5 Large AI21 Labs | 15 | 398B (94B active at inference time) | 256k | $3.5 | - | 🤗 | | View |
Gemma 3 4B Instruct Google | 15 | 4.3B | 128k | - | 52 | 🤗 | | View |
Hermes 3 - Llama-3.1 70B Nous Research | 15 | 70.6B | 128k | $0.3 | 21 | 🤗 | | View |
DeepSeek-Coder-V2 DeepSeek | 15 | 236B (21B active at inference time) | 128k | - | - | 🤗 | - | View |
Qwen3 1.7B (Non-reasoning) Alibaba | 14 | 2.03B | 32.0k | $0.2 | 116 | 🤗 | | View |
Jamba 1.6 Large AI21 Labs | 14 | 398B (94B active at inference time) | 256k | $3.5 | 44 | 🤗 | | View |
Qwen3 0.6B (Reasoning) Alibaba | 14 | 0.752B | 32.0k | $0.4 | 205 | 🤗 | | View |
Phi-3 Mini Instruct 3.8B Microsoft Azure | 13 | 3.8B | 4.10k | $0.2 | 69 | 🤗 | | View |
Llama 3 Instruct 70B Meta | 13 | 70B | 8.19k | $0.8 | 41 | 🤗 | +5 more | View |
Mistral Small (Sep '24) Mistral | 13 | 22B | 32.8k | $0.3 | 96 | 🤗 | | View |
Gemma 3n E4B Instruct Preview (May '25) Google | 13 | 8.39B (4B active at inference time) | 32.0k | - | - | 🤗 | - | View |
Phi-4 Multimodal Instruct Microsoft Azure | 12 | 5.6B | 128k | - | 18 | 🤗 | | View |
Qwen2.5 Coder Instruct 7B Alibaba | 12 | 7.62B | 131k | - | - | 🤗 | - | View |
 LFM2 2.6B Liquid AI | 12 | 2.57B | 32.8k | - | - | 🤗 | - | View |
Mixtral 8x22B Instruct Mistral | 12 | 141B (39B active at inference time) | 65.4k | $3.0 | 43 | 🤗 | | View |
Llama 2 Chat 7B Meta | 11 | 7B | 4.10k | $0.1 | 112 | 🤗 | | View |
Llama 3.2 Instruct 3B Meta | 11 | 3B | 128k | $0.0 | 110 | 🤗 | ? +3 more | View |
Qwen3 0.6B (Non-reasoning) Alibaba | 11 | 0.752B | 32.0k | $0.2 | 192 | 🤗 | | View |
Qwen1.5 Chat 110B Alibaba | 11 | 110B | 32.0k | - | - | 🤗 | - | View |
Phi-3 Medium Instruct 14B Microsoft Azure | 10 | 14B | 128k | $0.3 | 43 | 🤗 | | View |
Pixtral 12B (2409) Mistral | 9 | 12B | 128k | $0.1 | 156 | 🤗 | | View |
DeepSeek R1 Distill Qwen 1.5B DeepSeek | 9 | 1.5B | 128k | - | - | 🤗 | - | View |
DeepSeek-V2-Chat DeepSeek | 9 | 236B (21B active at inference time) | 128k | - | - | 🤗 | - | View |
LFM2 1.2B Liquid AI | 8 | 1.17B | 32.8k | - | - | 🤗 | ? | View |
Gemma 3n E2B Instruct Google | 8 | 5.98B (2.0B active at inference time) | 32.0k | - | 56 | 🤗 | | View |
Ministral 8B Mistral | 8 | 8B | 128k | $0.1 | 193 | 🤗 | | View |
Gemma 2 9B Google | 8 | 9B | 8.19k | $0.2 | - | 🤗 | | View |
 Arctic Instruct Snowflake | 8 | 480B (17B active at inference time) | 4.00k | - | - | 🤗 | - | View |
Qwen Chat 72B Alibaba | 8 | 72B | 33.8k | $1.0 | - | 🤗 | | View |
Llama 3.2 Instruct 1B Meta | 7 | 1B | 128k | $0.1 | 72 | 🤗 | | View |
Command-R+ (Aug '24) Cohere | 7 | 104B | 128k | $4.4 | 27 | 🤗 | | View |
Llama 3 Instruct 8B Meta | 7 | 8B | 8.19k | $0.1 | 65 | 🤗 | ? +2 more | View |
Gemma 3 1B Instruct Google | 6 | 1B | 32.0k | - | - | 🤗 | ? | View |
DeepSeek Coder V2 Lite Instruct DeepSeek | 6 | 16B (2.4B active at inference time) | 128k | - | - | 🤗 | - | View |
Codestral (May '24) Mistral | 6 | 22B | 32.8k | - | - | 🤗 | - | View |
Aya Expanse 32B Cohere | 6 | 32B | 128k | $0.8 | 70 | 🤗 | | View |
Llama 2 Chat 70B Meta | 6 | 70B | 4.10k | - | - | 🤗 | - | View |
DeepSeek LLM 67B Chat (V1) DeepSeek | 6 | 7B | 4.10k | - | - | 🤗 | - | View |
Llama 2 Chat 13B Meta | 6 | 13B | 4.10k | - | - | 🤗 | - | View |
Command-R+ (Apr '24) Cohere | 5 | 104B | 128k | $6.0 | - | 🤗 | | View |
 OpenChat 3.5 (1210) OpenChat | 5 | 7B | 8.19k | - | - | 🤗 | - | View |
DBRX Instruct Databricks | 5 | 132B (36B active at inference time) | 32.8k | - | - | 🤗 | - | View |
Mistral NeMo Mistral | 5 | 12B | 128k | $0.1 | 144 | 🤗 | +1 more | View |
Jamba 1.5 Mini AI21 Labs | 4 | 52B (12B active at inference time) | 256k | $0.3 | - | 🤗 | | View |
Jamba 1.7 Mini AI21 Labs | 4 | 52B (12.0B active at inference time) | 258k | $0.3 | 148 | 🤗 | | View |
Jamba 1.6 Mini AI21 Labs | 3 | 52B (12B active at inference time) | 256k | $0.3 | 151 | 🤗 | | View |
Mixtral 8x7B Instruct Mistral | 3 | 46.7B (12.9B active at inference time) | 32.8k | $0.7 | 62 | 🤗 | ? +2 more | View |
DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) Nous Research | 2 | 8B | 128k | - | - | 🤗 | - | View |
Aya Expanse 8B Cohere | 2 | 8B | 8.00k | $0.8 | 83 | 🤗 | | View |
Codestral-Mamba Mistral | 1 | 7B | 256k | - | - | 🤗 | - | View |
Mistral 7B Instruct Mistral | 1 | 7B | 8.19k | $0.3 | 118 | 🤗 | +3 more | View |
Command-R (Aug '24) Cohere | 1 | 32B | 128k | $0.3 | 72 | 🤗 | | View |
Command-R (Mar '24) Cohere | 1 | 35B | 128k | $0.8 | - | 🤗 | | View |