Open source AI models
375 open-weight models you can download or call through any host, ranked by quality.
/
 GLM-4.7-FlashZhipu AIMITJSON+2 | $0.40/1Mvia Amazon Bedrock | 200K | $0.07/1M | $0.40/1M | 7K | Jan 19 | |
 Qwen3.5 4BAlibabaApache 2.0Prompt caching+3 | $0.07/1Mvia EmpirioLabs AI | 262K | $0.04/1M | $0.07/1M | 7K | Feb 27 | |
| GLM-5.1Zhipu AIMITPrompt caching+3 | $1.44/1Mvia APIMart | 200K | $1.40/1M | $4.40/1M | 7K | Apr 3 | |
 Ling 3.0 Flash (free)InclusionaiMITPrompt caching+2 | $0.063/1Mvia OpenRouter | 262K | $0.021/1M | $0.063/1M | 6K | Aug 2 | |
| GLM-4.7Zhipu AIMITPrompt caching+3 | $1.10/1Mvia AIHubMix | 200K | $0.60/1M | $2.20/1M | 6K | Dec 22, 2025 | |
| GLM-4.5Zhipu AIMITPrompt caching+3 | $1.16/1Mvia ZenMux | 131K | $0.60/1M | $2.20/1M | 6K | Jul 20, 2025 | |
| GLM-4.6Zhipu AIMITPrompt caching+3 | $1.10/1Mvia AIHubMix | 200K | $0.60/1M | $2.20/1M | 6K | Sep 29, 2025 | |
 Gemma 4GoogleApache 2.0JSON+3 | $0.30/1Mvia FastRouter | 262K | $0.10/1M | $0.30/1M | 6K | Mar 11 | |
| Qwen3-VL-8B-InstructAlibabaApache 2.0Prompt caching+3 | $0.278/1Mvia SiliconFlow | 262K | $0.069/1M | $0.278/1M | 6K | Oct 11, 2025 | |
| Kimi K2.7 Code HighSpeedMoonshot AIOpen weights · with termsPrompt caching+4 | $2.76/1Mvia Kenari | 262K | $0.903/1M | $3.75/1M | 6K | Jun 11 | |
| LFM2.5-24B-A2BLiquid AIOpen weights · with termsPrompt caching+3 | $0.70/1Mvia OpenRouter | 128K | $0.05/1M | $0.70/1M | 6K | Oct 12 | |
| Qwen3-VL-8B-ThinkingAlibabaApache 2.0JSON+3 | $0.694/1Mvia SiliconFlow | 131K | $0.069/1M | $0.694/1M | 6K | Oct 11, 2025 | |
| Qwen2.5-Coder-7BAlibabaApache 2.0 | $0.03/1Mvia Hugging Face | — | $0.01/1M | $0.03/1M | 6K | Sep 16, 2024 | |
| Hermes 3 405B InstructNous ResearchLlama 3JSON | $1/1Mvia OpenRouter | 131K | $1/1M | $1/1M | 5K | Aug 13, 2024 | |
 MiMo-V2.5-ProXiaomiMITPrompt caching+3 | $0.609/1Mvia OpenRouter | 1.05M | $0.304/1M | $0.609/1M | 5K | Apr 27 | |
| Qwen3 30B A3BAlibabaApache 2.0JSON+2 | $0.29/1Mvia Helicone | 41K | $0.08/1M | $0.29/1M | 5K | Apr 27, 2025 | |
| Qwen3-Omni-30B-A3B-CaptionerAlibabaOpen weights · permissive | $0.389/1Mvia SiliconFlow | — | $0.097/1M | $0.389/1M | 5K | Sep 15, 2025 | |
 Llama-2-13b-chat-hfMetaLlama 2 | $0.50/1Mvia Replicate | — | $0.10/1M | $0.50/1M | 5K | Jul 9, 2023 | |
| Llama-3.3-70B-InstructMetaLlama 3.3JSON+1 | $0.32/1Mvia Deep Infra | 131K | $0.10/1M | $0.32/1M | 5K | Nov 26, 2024 | |
| Qwen3.6 27BAlibabaApache 2.0JSON+3 | $2/1Mvia OpenRouter | 262K | $0.60/1M | $3.60/1M | 4K | Apr 21 | |
| Gemma 3 4B ITGoogleGemmaPrompt caching+2 | $0.05/1Mvia OpenRouter | 131K | $0.04/1M | $0.05/1M | 4K | Feb 20, 2025 | |
 DeepSeek V4 Flash 0731DeepSeekMITPrompt caching+3 | $0.133/1Mvia OpenRouter | 1.05M | $0.044/1M | $0.133/1M | 4K | Jul 31 | |
| Qwen2 57B A14B InstructAlibabaApache 2.0 | $0.24/1Mvia AIHubMix | — | $0.24/1M | $0.24/1M | 4K | Jun 4, 2024 | |
| LongCat-2.0MeituanMITPrompt caching+2 | $1.20/1Mvia Atlas Cloud | 1.05M | $0.30/1M | $1.20/1M | 4K | Jul 5 | |
| Qwen3 0.6BAlibabaApache 2.0 | $0.46/1Mvia AIHubMix | — | $0.046/1M | $0.46/1M | 4K | Apr 27, 2025 | |
| Qwen3.5 0.8BAlibabaApache 2.0 | $0.08/1Mvia Prime Inference | — | $0.04/1M | $0.08/1M | 3K | Feb 28 | |
| Qwen3 14BAlibabaApache 2.0JSON+2 | $0.20/1Mvia Hugging Face | 41K | $0.07/1M | $0.20/1M | 3K | Apr 27, 2025 | |
| Qwen3.5 397B-A17BAlibabaApache 2.0JSON+3 | $0.986/1Mvia AIHubMix | 262K | $0.172/1M | $1.03/1M | 3K | Feb 16 | |
| Qwen3.5 2BAlibabaApache 2.0JSON+3 | $0.10/1Mvia Requesty | 262K | $0.02/1M | $0.10/1M | 3K | Feb 28 | |
| Qwen2.5-7B-InstructAlibabaApache 2.0JSON+1 | $0.05/1Mvia SiliconFlow | 33K | $0.05/1M | $0.05/1M | 3K | Sep 16, 2024 | |
| Qwen3 Coder 480B A35B InstructAlibabaApache 2.0Prompt caching+2 | $0.95/1Mvia Helicone | 262K | $0.22/1M | $0.95/1M | 3K | Jul 22, 2025 | |
| Llama 3.1 8B InstructMetaLlama 3.1JSON+1 | $0.05/1Mvia Helicone | 131K | $0.02/1M | $0.05/1M | 3K | Jul 18, 2024 | |
| Qwen3-Coder 30B-A3B InstructAlibabaApache 2.0JSON+1 | $0.27/1Mvia DevPass (LLM Gateway) | 262K | $0.07/1M | $0.27/1M | 3K | Jul 31, 2025 | |
| Qwen3.5 35B-A3BAlibabaApache 2.0JSON+3 | $0.451/1Mvia AIHubMix | 262K | $0.25/1M | $2/1M | 3K | Feb 24 | |
| Gemma 3 27BGoogleGemmaJSON+2 | $0.16/1Mvia Deep Infra | 131K | $0.08/1M | $0.16/1M | 3K | Mar 1, 2025 | |
| GLM-4.5-AirZhipu AIMITPrompt caching+3 | $0.291/1Mvia ZenMux | 131K | $0.20/1M | $1.10/1M | 3K | Jul 20, 2025 | |
| DeepSeek R1 Distill Qwen 7BDeepSeekMIT | $0.01/1Mvia AIHubMix | — | $0.01/1M | $0.01/1M | 3K | Jan 20, 2025 | |
| Gemma 3 1B ItGoogleGemma | $0.20/1Mvia AIHubMix | — | $0.20/1M | $0.20/1M | 3K | Mar 10, 2025 | |
| Gemma 4 26B A4B ITGoogleApache 2.0Prompt caching+4 | $0.225/1Mvia OpenRouter | 262K | $0.15/1M | $0.60/1M | 3K | Mar 11 | |
| Qwen2.5 3B InstructAlibabaNon-commercial | $0.80/1Mvia AIHubMix | — | $0.40/1M | $0.80/1M | 2K | Sep 17, 2024 | |
| Qwen3 32BAlibabaApache 2.0JSON+2 | $0.25/1Mvia Hugging Face | 131K | $0.08/1M | $0.25/1M | 2K | Apr 27, 2025 | |
| FLUX.1-devBlack Forest LabsNon-commercial | $0.025/MPvia Fal | — | — | $0.025/MP | 7K | Jul 31, 2024 | |
| Qwen Image LayeredAlibabaApache 2.0 | $0.05/imgvia Fal | — | — | $0.05/img | 5K | Dec 17, 2025 | |
| Qwen Image 2512AlibabaApache 2.0 | $0.02/imgvia FastRouter | 4K | — | $0.02/img | 4K | Dec 30, 2025 | |
| FLUX 2 DevBlack Forest LabsNon-commercial | $0.012/runvia WaveSpeedAI | — | — | $0.012/run | 3K | Nov 22, 2025 | |
 Hunyuan Video 1.5TencentOpen weights · with terms | $0.04/runvia WaveSpeedAI | — | — | $0.04/run | 6K | Nov 18, 2025 | |
| LTX 2LightricksOpen weights · with terms | $0.10/runvia WaveSpeedAI | — | — | $0.10/run | 6K | Jan 3 | |
| LongCat VideoMeituanMIT | $0.025/secvia Fal | — | — | $0.025/sec | 4K | Oct 24, 2025 | |
| Qwen 3 Embedding 4BAlibabaApache 2.0 | $0.019/1Mvia SiliconFlow | 33K | — | $0.019/1M | 5K | Jun 3, 2025 | |
 Nemotron 3.5 Embed 1BNVIDIAOpen weights · permissive | Free/1Mvia Nvidia | — | Free/1M | Free/1M | 4K | Oct 7 |
Frequently asked questions
Ranked here by quality. Open models now trail the best closed ones by little, and the hosts that serve them often charge much less.
Most "open source" models publish their weights under a license, but not their training data. We list every model whose weights you can download.
On your own hardware, or through inference hosts that charge per token. Open a model's providers to compare each host's price, speed and uptime.
Can't find what you're looking for? Contact us