self-hosted
Self-hostable open-weight models
Models from Llama, Mistral, Qwen, DeepSeek, and other vendors with downloadable weights.
| # | Model | Score | In / 1M | Out / 1M | Context | |
|---|---|---|---|---|---|---|
| 1 | Google: Gemma 4 26B A4B (free)google/gemma-4-26b-a4b-it:free | 100 | Free | Free | 262,144 | Try → |
| 2 | Google: Gemma 4 26B A4B google/gemma-4-26b-a4b-it | 100 | $0.070 | $0.350 | 262,144 | Try → |
| 3 | Google: Gemma 4 31B (free)google/gemma-4-31b-it:free | 100 | Free | Free | 262,144 | Try → |
| 4 | Google: Gemma 4 31Bgoogle/gemma-4-31b-it | 100 | $0.130 | $0.380 | 262,144 | Try → |
| 5 | Qwen: Qwen3.6 Plusqwen/qwen3.6-plus | 100 | $0.325 | $1.950 | 1,000,000 | Try → |
| 6 | Mistral: Mistral Small 4mistralai/mistral-small-2603 | 100 | $0.150 | $0.600 | 262,144 | Try → |
| 7 | Qwen: Qwen3.5-9Bqwen/qwen3.5-9b | 100 | $0.100 | $0.150 | 262,144 | Try → |
| 8 | Google: Gemini 3.1 Flash Lite Previewgoogle/gemini-3.1-flash-lite-preview | 100 | $0.250 | $1.500 | 1,048,576 | Try → |
| 9 | Qwen: Qwen3.5-35B-A3Bqwen/qwen3.5-35b-a3b | 100 | $0.163 | $1.300 | 262,144 | Try → |
| 10 | Qwen: Qwen3.5-27Bqwen/qwen3.5-27b | 100 | $0.195 | $1.560 | 262,144 | Try → |
| 11 | Qwen: Qwen3.5-122B-A10Bqwen/qwen3.5-122b-a10b | 100 | $0.260 | $2.080 | 262,144 | Try → |
| 12 | Qwen: Qwen3.5-Flashqwen/qwen3.5-flash-02-23 | 100 | $0.065 | $0.260 | 1,000,000 | Try → |
| 13 | Google: Gemini 3.1 Pro Preview Custom Toolsgoogle/gemini-3.1-pro-preview-customtools | 100 | $2.000 | $12.000 | 1,048,576 | Try → |
| 14 | Google: Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview | 100 | $2.000 | $12.000 | 1,048,576 | Try → |
| 15 | Qwen: Qwen3.5 Plus 2026-02-15qwen/qwen3.5-plus-02-15 | 100 | $0.260 | $1.560 | 1,000,000 | Try → |
| 16 | Qwen: Qwen3.5 397B A17Bqwen/qwen3.5-397b-a17b | 100 | $0.390 | $2.340 | 262,144 | Try → |
| 17 | Google: Gemini 3 Flash Previewgoogle/gemini-3-flash-preview | 100 | $0.500 | $3.000 | 1,048,576 | Try → |
| 18 | Google: Gemini 2.5 Flash Lite Preview 09-2025google/gemini-2.5-flash-lite-preview-09-2025 | 100 | $0.100 | $0.400 | 1,048,576 | Try → |
| 19 | Google: Gemini 2.5 Flash Litegoogle/gemini-2.5-flash-lite | 100 | $0.100 | $0.400 | 1,048,576 | Try → |
| 20 | Google: Gemini 2.5 Flashgoogle/gemini-2.5-flash | 100 | $0.300 | $2.500 | 1,048,576 | Try → |
| 21 | Google: Gemini 2.5 Progoogle/gemini-2.5-pro | 100 | $1.250 | $10.000 | 1,048,576 | Try → |
| 22 | Google: Gemini 2.5 Pro Preview 06-05google/gemini-2.5-pro-preview | 100 | $1.250 | $10.000 | 1,048,576 | Try → |
| 23 | Google: Gemini 2.5 Pro Preview 05-06google/gemini-2.5-pro-preview-05-06 | 100 | $1.250 | $10.000 | 1,048,576 | Try → |
| 24 | Google: Gemini 2.0 Flash Litegoogle/gemini-2.0-flash-lite-001 | 100 | $0.075 | $0.300 | 1,048,576 | Try → |
| 25 | Google: Gemini 2.0 Flashgoogle/gemini-2.0-flash-001 | 100 | $0.100 | $0.400 | 1,048,576 | Try → |
| 26 | NVIDIA: Nemotron 3 Nano 30B A3Bnvidia/nemotron-3-nano-30b-a3b | 99 | $0.050 | $0.200 | 262,144 | Try → |
| 27 | Mistral: Ministral 3 14B 2512mistralai/ministral-14b-2512 | 99 | $0.200 | $0.200 | 262,144 | Try → |
| 28 | Mistral: Ministral 3 8B 2512mistralai/ministral-8b-2512 | 99 | $0.150 | $0.150 | 262,144 | Try → |
| 29 | Meta: Llama 4 Scoutmeta-llama/llama-4-scout | 99 | $0.080 | $0.300 | 327,680 | Try → |
| 30 | NVIDIA: Nemotron 3 Supernvidia/nemotron-3-super-120b-a12b | 99 | $0.090 | $0.450 | 262,144 | Try → |
| 31 | Qwen: Qwen3 235B A22B Thinking 2507qwen/qwen3-235b-a22b-thinking-2507 | 99 | $0.130 | $0.600 | 262,144 | Try → |
| 32 | Qwen: Qwen3 VL 235B A22B Instructqwen/qwen3-vl-235b-a22b-instruct | 99 | $0.200 | $0.880 | 262,144 | Try → |
| 33 | Qwen: Qwen Plus 0728 (thinking)qwen/qwen-plus-2025-07-28:thinking | 99 | $0.260 | $0.780 | 1,000,000 | Try → |
| 34 | Mistral: Mistral Large 3 2512mistralai/mistral-large-2512 | 99 | $0.500 | $1.500 | 262,144 | Try → |
| 35 | Qwen: Qwen3 Max Thinkingqwen/qwen3-max-thinking | 97 | $0.780 | $3.900 | 262,144 | Try → |
| 36 | Qwen: Qwen3 VL 8B Thinkingqwen/qwen3-vl-8b-thinking | 95 | $0.117 | $1.365 | 131,072 | Try → |
| 37 | Qwen: Qwen3 VL 30B A3B Thinkingqwen/qwen3-vl-30b-a3b-thinking | 95 | $0.130 | $1.560 | 131,072 | Try → |
| 38 | DeepSeek: DeepSeek V3.2 Expdeepseek/deepseek-v3.2-exp | 95 | $0.270 | $0.410 | 163,840 | Try → |
| 39 | Qwen: Qwen3 VL 235B A22B Thinkingqwen/qwen3-vl-235b-a22b-thinking | 95 | $0.260 | $2.600 | 131,072 | Try → |
| 40 | DeepSeek: DeepSeek V3.1 Terminusdeepseek/deepseek-v3.1-terminus | 95 | $0.210 | $0.790 | 163,840 | Try → |
| 41 | Qwen: Qwen3 235B A22B Instruct 2507qwen/qwen3-235b-a22b-2507 | 94 | $0.071 | $0.100 | 262,144 | Try → |
| 42 | Qwen: Qwen3 30B A3B Instruct 2507qwen/qwen3-30b-a3b-instruct-2507 | 94 | $0.090 | $0.300 | 262,144 | Try → |
| 43 | Qwen: Qwen3 Coder Nextqwen/qwen3-coder-next | 94 | $0.150 | $0.800 | 262,144 | Try → |
| 44 | Qwen: Qwen Plus 0728qwen/qwen-plus-2025-07-28 | 94 | $0.260 | $0.780 | 1,000,000 | Try → |
| 45 | Qwen: Qwen-Plusqwen/qwen-plus | 94 | $0.260 | $0.780 | 1,000,000 | Try → |
| 46 | Qwen: Qwen3 Coder Flashqwen/qwen3-coder-flash | 94 | $0.195 | $0.975 | 1,000,000 | Try → |
| 47 | Qwen: Qwen3 Next 80B A3B Instructqwen/qwen3-next-80b-a3b-instruct | 94 | $0.090 | $1.100 | 262,144 | Try → |
| 48 | Mistral: Codestral 2508mistralai/codestral-2508 | 94 | $0.300 | $0.900 | 256,000 | Try → |
| 49 | Qwen: Qwen3 Coder 480B A35Bqwen/qwen3-coder | 94 | $0.220 | $1.000 | 262,144 | Try → |
| 50 | DeepSeek: R1 0528deepseek/deepseek-r1-0528 | 94 | $0.500 | $2.150 | 163,840 | Try → |