695 listed
Models
Language models for agents: hosted models with input and output prices per million tokens from OpenRouter (read daily; prices change often), and open-weights models from Hugging Face with license and downloads. Filter by tool calling, context window and price, or switch to the table to compare.
Showing 625–648 of 695
| Model | Provider | Input $/M | Output $/M | Context | Tools | Weights |
|---|---|---|---|---|---|---|
| Llama3-ChatQA-1.5-70B | NVIDIA | Self-host | — | — | — | Open |
| Meta-Llama-Guard-2-8B | Meta | Self-host | — | — | — | Open |
| MiniCPM4.1-8B | openbmb | Self-host | — | — | — | Open |
| Nous-Hermes-Llama2-13b | NousResearch | Self-host | — | — | — | Open |
| OLMo-7B | allenai | Self-host | — | — | — | Open |
| solar-pro-preview-instruct | upstage | Self-host | — | — | — | Open |
| starcoder | bigcode | Self-host | — | — | — | Open |
| xgen-7b-8k-base | Salesforce | Self-host | — | — | — | Open |
| Yarn-Mistral-7b-128k | NousResearch | Self-host | — | — | — | Open |
| Yi-34B-200K | 01-ai | Self-host | — | — | — | Open |
| CodeQwen1.5-7B-Chat | Qwen | Self-host | — | — | — | Open |
| Qwen-72B | Qwen | Self-host | — | — | — | Open |
| Qwen2.5-14B-Instruct-1M | Qwen | Self-host | — | — | — | Open |
| DeepSeek-Prover-V2-671B | deepseek-ai | Self-host | — | — | — | Open |
| Hunyuan-1.8B-Instruct | tencent | Self-host | — | — | — | Open |
| MistralLite | Amazon | Self-host | — | — | — | Open |
| santacoder | bigcode | Self-host | — | — | — | Open |
| starcoderbase | bigcode | Self-host | — | — | — | Open |
| CodeLlama-7b-hf | codellama | Self-host | — | — | — | Open |
| gpt-neo-125m | EleutherAI | Self-host | — | — | — | Open |
| gpt-neox-20b | EleutherAI | Self-host | — | — | — | Open |
| macbert4csc-base-chinese | shibing624 | Self-host | — | — | — | Open |
| pythia-160m-deduped | EleutherAI | Self-host | — | — | — | Open |
| pythia-70m-deduped | EleutherAI | Self-host | — | — | — | Open |
Prices are USD per million tokens from OpenRouter, read daily. Providers change prices often; check the model page before you rely on one.