{"data":[{"provider":"synthetic","display_name":"DeepSeek V4.1 Flash","always_on":true,"id":"syn:large:text","hugging_face_id":"deepseek-ai/DeepSeek-V4.1-Flash","name":"syn:large:text","reasoning_parameters":{"efforts":["none","low","high","xhigh","max"]},"description":"An efficient near-Fable model with vision.","input_modalities":["text","image"],"output_modalities":["text"],"context_length":524288,"max_output_length":65536,"pricing":{"prompt":"$0.0000006","completion":"$0.0000012","image":"0","request":"0","input_cache_reads":"$0.00000003","input_cache_writes":"0"},"created":1788912000,"quantization":"fp8","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"deepseek-ai/deepseek-v4.1-flash"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"GLM 4.7 Flash","always_on":true,"id":"syn:small:text","hugging_face_id":"zai-org/GLM-4.7-Flash","name":"syn:small:text","reasoning_parameters":{"efforts":["none","low","medium","high"]},"description":"The smaller, faster sibling of GLM 4.7, in the 30B parameter class, balancing performance and efficiency. Consumes rate limits more slowly.","input_modalities":["text"],"output_modalities":["text"],"context_length":196608,"max_output_length":65536,"pricing":{"prompt":"$0.0000001","completion":"$0.0000005","image":"0","request":"0","input_cache_reads":"$0.00000002","input_cache_writes":"0"},"created":1768780800,"quantization":"bf16","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"minimaxai/minimax-m2.5"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"Kimi K3","always_on":true,"id":"syn:large:vision","hugging_face_id":"moonshotai/Kimi-K3","name":"syn:large:vision","reasoning_parameters":{"efforts":["low","high","max"]},"description":"A massive 2.8T-parameter open-source model, on par with Claude Fable and GPT-5.6-Sol. Supports vision on uploaded images.","input_modalities":["text","image"],"output_modalities":["text"],"context_length":524288,"max_output_length":65536,"pricing":{"prompt":"$0.000003","completion":"$0.000015","image":"0","request":"0","input_cache_reads":"$0.00000045","input_cache_writes":"0"},"created":1785110400,"quantization":"mxfp4","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"moonshotai/kimi-k3"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"Qwen 3.8 27B","always_on":true,"id":"syn:small:vision","hugging_face_id":"Qwen/Qwen3.8-27B-FP8","name":"syn:small:vision","reasoning_parameters":{"efforts":["low","medium","xhigh"]},"description":"A small but capable coding and vision model from Qwen. Fast and consumes rate limits less quickly.","input_modalities":["text","image"],"output_modalities":["text"],"context_length":262144,"max_output_length":65536,"pricing":{"prompt":"$0.00000045","completion":"$0.0000022","image":"0","request":"0","input_cache_reads":"$0.00000009","input_cache_writes":"0"},"created":1786665600,"quantization":"fp8","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"qwen/qwen3.8-27b"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"GPT-OSS-120B","always_on":true,"id":"hf:openai/gpt-oss-120b","hugging_face_id":"openai/gpt-oss-120b","name":"openai/gpt-oss-120b","reasoning_parameters":{"efforts":["none","low","medium","high"]},"description":"OpenAI's largest open-source model. A strong reasoning model with great support for agentic tasks, and a good general-purpose LLM.","input_modalities":["text"],"output_modalities":["text"],"context_length":131072,"max_output_length":65536,"pricing":{"prompt":"$0.0000001","completion":"$0.0000001","image":"0","request":"0","input_cache_reads":"$0.00000002","input_cache_writes":"0"},"created":1754352000,"quantization":"mxfp4","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"openai/gpt-oss-120b"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"GLM 5.3 Flash","always_on":true,"id":"hf:zai-org/GLM-5.3-Flash","hugging_face_id":"zai-org/GLM-5.3-Flash","name":"zai-org/GLM-5.3-Flash","reasoning_parameters":{"efforts":["low","high","max"]},"description":"An extremely efficient, Opus-4.8-equivalent model with vision.","input_modalities":["text","image"],"output_modalities":["text"],"context_length":524288,"max_output_length":65536,"pricing":{"prompt":"$0.00000015","completion":"$0.0000005","image":"0","request":"0","input_cache_reads":"$0.00000004","input_cache_writes":"0"},"created":1787702400,"quantization":"fp8","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"zai-org/glm-5.3-flash"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"DeepSeek V4.1 Flash","always_on":true,"id":"hf:deepseek-ai/DeepSeek-V4.1-Flash","hugging_face_id":"deepseek-ai/DeepSeek-V4.1-Flash","name":"deepseek-ai/DeepSeek-V4.1-Flash","reasoning_parameters":{"efforts":["none","low","high","xhigh","max"]},"description":"An efficient near-Fable model with vision.","input_modalities":["text","image"],"output_modalities":["text"],"context_length":524288,"max_output_length":65536,"pricing":{"prompt":"$0.0000006","completion":"$0.0000012","image":"0","request":"0","input_cache_reads":"$0.00000003","input_cache_writes":"0"},"created":1788912000,"quantization":"fp8","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"deepseek-ai/deepseek-v4.1-flash"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"Kimi K3","always_on":true,"id":"hf:moonshotai/Kimi-K3","hugging_face_id":"moonshotai/Kimi-K3","name":"moonshotai/Kimi-K3","reasoning_parameters":{"efforts":["low","high","max"]},"description":"A massive 2.8T-parameter open-source model, on par with Claude Fable and GPT-5.6-Sol. Supports vision on uploaded images.","input_modalities":["text","image"],"output_modalities":["text"],"context_length":524288,"max_output_length":65536,"pricing":{"prompt":"$0.000003","completion":"$0.000015","image":"0","request":"0","input_cache_reads":"$0.00000045","input_cache_writes":"0"},"created":1785110400,"quantization":"mxfp4","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"moonshotai/kimi-k3"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"Qwen 3.8 27B","always_on":true,"id":"hf:Qwen/Qwen3.8-27B","hugging_face_id":"Qwen/Qwen3.8-27B-FP8","name":"Qwen/Qwen3.8-27B","reasoning_parameters":{"efforts":["low","medium","xhigh"]},"description":"A small but capable coding and vision model from Qwen. Fast and consumes rate limits less quickly.","input_modalities":["text","image"],"output_modalities":["text"],"context_length":262144,"max_output_length":65536,"pricing":{"prompt":"$0.00000045","completion":"$0.0000022","image":"0","request":"0","input_cache_reads":"$0.00000009","input_cache_writes":"0"},"created":1786665600,"quantization":"fp8","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"qwen/qwen3.8-27b"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"GLM 4.7 Flash","always_on":true,"id":"hf:zai-org/GLM-4.7-Flash","hugging_face_id":"zai-org/GLM-4.7-Flash","name":"zai-org/GLM-4.7-Flash","reasoning_parameters":{"efforts":["none","low","medium","high"]},"description":"The smaller, faster sibling of GLM 4.7, in the 30B parameter class, balancing performance and efficiency. Consumes rate limits more slowly.","input_modalities":["text"],"output_modalities":["text"],"context_length":196608,"max_output_length":65536,"pricing":{"prompt":"$0.0000001","completion":"$0.0000005","image":"0","request":"0","input_cache_reads":"$0.00000002","input_cache_writes":"0"},"created":1768780800,"quantization":"bf16","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"minimaxai/minimax-m2.5"},"datacenters":[{"country_code":"US"}]},{"provider":"synthetic","display_name":"Nemotron 3 Super 120B","always_on":true,"id":"hf:nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4","hugging_face_id":"nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8","name":"nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4","reasoning_parameters":{"efforts":["none","low","medium","high"]},"description":"A fast, 120b-parameter model with good support for coding agents. Consumes rate limits more slowly.","input_modalities":["text"],"output_modalities":["text"],"context_length":262144,"max_output_length":65536,"pricing":{"prompt":"$0.0000003","completion":"$0.000001","image":"0","request":"0","input_cache_reads":"$0.00000006","input_cache_writes":"0"},"created":1773187200,"quantization":"fp8","supported_sampling_parameters":["temperature","top_k","top_p","repetition_penalty","frequency_penalty","presence_penalty","stop","seed"],"supported_features":["tools","json_mode","structured_outputs","reasoning"],"openrouter":{"slug":"nvidia/nemotron-3-super-120b-a12b"},"datacenters":[{"country_code":"US"}]}]}