Models
Models available through the service, with their capabilities and context window.
The list is read from the service catalog when the page is requested. Its contents may change.
Models in total: 468
General purpose
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| aion-2.0 | 131K | text | text | prompt cachingreasoningreasoning requiredtool calling |
| aion-3.0 | 131K | text | text | prompt cachingreasoningreasoning requiredtool calling |
| aion-3.0-mini | 131K | text | text | prompt cachingreasoningreasoning requiredtool calling |
| aion-rp-llama-3.1-8b | 33K | text | text | |
| claude-3-haiku | 200K | images, text | text | prompt cachingtool callingvisionweb search |
| claude-fable-5 | 1M | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| command-a | 256K | text | text | reasoningstructured outputstool calling |
| command-r | 128K | text | text | structured outputstool calling |
| command-r-plus | 128K | text | text | structured outputstool calling |
| command-r7b | 128K | text | text | structured outputs |
| cydonia-24b-v4.1 | 131K | text | text | logprobsprompt cachingstructured outputs |
| deepseek-v4-flash | 800K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| deepseek-v4-flash-vision-exp | 1.0M | images, text | text | logprobsprompt cachingreasoningtool callingvision |
| deepseek-v4-pro | 800K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| dolphin-mistral-24b-venice-edition | 128K | text | text | |
| dots-3-note | 512K | images, text | text | reasoningstructured outputstool callingvision |
| ernie-4.5-vl-424b-a47b | 123K | images, text | text | reasoningvision |
| fugu-ultra | 1M | images, text | text | prompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gemini-2.5-flash | 120K | audio, files, images, text, video | text | audio inputfile inputlogprobsprompt cachingreasoningstructured outputstool callingvideo inputvisionweb search |
| gemini-2.5-flash-lite | 120K | audio, files, images, text, video | text | audio inputfile inputprompt cachingreasoningstructured outputstool callingvideo inputvisionweb search |
| gemini-2.5-pro | 120K | audio, files, images, text, video | text | audio inputfile inputlogprobsprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| gemini-3-flash | 120K | audio, files, images, text, video | text | audio inputfile inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvisionweb search |
| gemini-3.1-flash-lite | 200K | audio, files, images, text, video | text | audio inputfile inputlogprobsprompt cachingreasoningstructured outputstool callingvideo inputvisionweb search |
| gemini-3.1-pro | 120K | audio, files, images, text, video | text | audio inputfile inputparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| gemini-3.1-pro-preview-customtools | 1.0M | audio, files, images, text, video | text | audio inputfile inputprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| gemini-3.5-flash | 1.0M | audio, files, images, text, video | text | audio inputfile inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| gemini-3.5-flash-lite | 1.0M | audio, files, images, text, video | text | audio inputfile inputparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| gemini-3.6-flash | 1.0M | audio, files, images, text, video | text | audio inputfile inputparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| gemini-3.7-flash | 1.0M | audio, files, images, text, video | text | audio inputfile inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| gemini-3.7-flash:batch | 1.0M | audio, files, images, text, video | text | audio inputfile inputprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| gemma-2-27b-it | 8K | text | text | structured outputs |
| gemma-3-12b-it | 131K | images, text | text | structured outputstool callingvision |
| gemma-3-27b-it | 128K | images, text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvision |
| gemma-3-4b-it | 131K | images, text | text | structured outputsvision |
| gemma-3n-e4b-it | 33K | text | text | structured outputs |
| gemma-4-26b-a4b-it | 262K | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvisionweb search |
| gemma-4-31b-it | 262K | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvisionweb search |
| glm-4.5 | 131K | text | text | logprobsprompt cachingreasoningtool calling |
| glm-4.5-air | 131K | text | text | prompt cachingreasoningtool calling |
| glm-4.5v | 66K | images, text | text | prompt cachingreasoningtool callingvision |
| glm-4.6 | 180K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| glm-4.6v | 131K | images, text, video | text | prompt cachingreasoningtool callingvideo inputvision |
| glm-4.7 | 180K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| glm-4.7-flash | 203K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| glm-5 | 203K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| glm-5-turbo | 203K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningtool calling |
| glm-5.1 | 180K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| glm-5.2 | 250K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| glm-5.3 | 1.0M | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredtool calling |
| glm-5v-turbo | 203K | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningtool callingvideo inputvision |
| gpt-3.5-turbo | 16K | text | text | logprobsstructured outputstool calling |
| gpt-3.5-turbo-16k | 16K | text | text | logprobsstructured outputstool calling |
| gpt-3.5-turbo-instruct | 4K | text | text | logprobsstructured outputs |
| gpt-4 | 8K | files, text | text | file inputlogprobsparallel tool callspredicted outputsstructured outputstool callingweb search |
| gpt-4-turbo | 128K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsstructured outputstool callingvisionweb search |
| gpt-4.1 | 64K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingstructured outputstool callingvisionweb search |
| gpt-4.1-mini | 128K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingstructured outputstool callingvisionweb search |
| gpt-4.1-nano | 120K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingstructured outputstool callingvisionweb search |
| gpt-4o | 120K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-4o-mini | 120K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5 | 380K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5-mini | 380K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5-nano | 380K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5-pro | 400K | files, images, text | text | file inputreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5.1 | 380K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.1-chat | 128K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningtool callingvisionweb search |
| gpt-5.2 | 380K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.2-chat | 128K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.2-pro | 400K | files, images, text | text | file inputparallel tool callspredicted outputsreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5.3-chat | 128K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningtool callingvisionweb search |
| gpt-5.4 | 250K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.4-mini | 380K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.4-nano | 380K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.4-pro | 1.1M | files, images, text | text | file inputparallel tool callspredicted outputsreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5.5 | 250K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.5-pro | 1.1M | files, images, text | text | file inputparallel tool callspredicted outputsreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5.6-luna | 256K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.6-luna-pro | 1.1M | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.6-sol | 256K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.6-sol-pro | 1.1M | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.6-terra | 256K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.6-terra-pro | 1.1M | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-audio | 128K | audio, text | audio, text | audio inputaudio outputlogprobsstructured outputstool calling |
| gpt-audio-mini | 128K | audio, text | audio, text | audio inputaudio outputlogprobsstructured outputstool calling |
| gpt-oss-120b | 64K | files, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool calling |
| gpt-oss-20b | 131K | files, text | text | file inputlogprobsparallel tool callsprompt cachingreasoningreasoning requiredstructured outputstool calling |
| gpt-oss-safeguard-20b | 131K | text | text | prompt cachingreasoningreasoning requiredstructured outputstool calling |
| granite-4.0-h-micro | 131K | text | text | logprobs |
| granite-4.1-8b | 131K | text | text | logprobsprompt cachingstructured outputstool calling |
| grok-4 | 250K | text | text | |
| grok-4.20 | 2M | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| grok-4.20-beta | 2M | images, text | text | parallel tool callsprompt cachingreasoningtool callingvisionweb search |
| grok-4.20-multi-agent | 2M | files, images, text | text | file inputlogprobsprompt cachingreasoningreasoning requiredstructured outputsvisionweb search |
| grok-4.3 | 1M | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| grok-4.5 | 500K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| grok-4.6 | 500K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| grok-build-0.1 | 256K | files, images, text | text | file inputlogprobsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| hermes-3-llama-3.1-405b | 131K | text | text | structured outputs |
| hermes-3-llama-3.1-70b | 131K | text | text | structured outputs |
| hermes-4-405b | 131K | text | text | reasoning |
| hermes-4-70b | 131K | text | text | reasoning |
| hunyuan-a13b-instruct | 131K | text | text | reasoningstructured outputs |
| hy-mt2-1.8b | 8K | text | text | |
| hy-mt2-30b-a3b | 8K | text | text | structured outputs |
| hy-mt2-7b | 8K | text | text | structured outputs |
| hy3 | 262K | text | text | prompt cachingreasoningstructured outputstool calling |
| inkling | 1.0M | audio, images, text | text | audio inputprompt cachingreasoningtool callingvision |
| inkling-small | 1.0M | audio, images, text | text | audio inputlogprobsprompt cachingreasoningtool callingvision |
| kimi-k2 | 131K | text | text | logprobstool calling |
| kimi-k2-thinking | 262K | text | text | logprobsprompt cachingreasoningreasoning requiredstructured outputstool calling |
| kimi-k2.5 | 262K | images, text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvision |
| kimi-k2.6 | 262K | images, text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvision |
| kimi-k3 | 1M | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvision |
| l3-lunaris-8b | 8K | text | text | logprobsstructured outputs |
| l3.1-euryale-70b | 131K | text | text | structured outputstool calling |
| l3.3-euryale-70b | 131K | text | text | logprobsstructured outputs |
| laguna-s-2.1 | 1.0M | text | text | prompt cachingreasoningtool calling |
| laguna-xs-2.1 | 262K | text | text | prompt cachingreasoningtool calling |
| lfm-2.5-2.6b | 66K | text | text | logprobsreasoningreasoning requiredstructured outputstool calling |
| ling-2.6-1t | 262K | text | text | logprobsprompt cachingstructured outputstool calling |
| ling-2.6-flash | 262K | text | text | logprobsprompt cachingstructured outputstool calling |
| ling-3.0-flash | 262K | text | text | logprobsprompt cachingreasoningtool calling |
| llama-3.1-70b-instruct | 131K | text | text | logprobsreasoningstructured outputstool calling |
| llama-3.1-8b-instruct | 16K | text | text | logprobsprompt cachingreasoningstructured outputstool calling |
| llama-3.2-1b-instruct | 60K | text | text | logprobsreasoningtool calling |
| llama-3.2-3b-instruct | 131K | text | text | logprobsreasoningstructured outputstool calling |
| llama-3.3-70b-instruct | 131K | text | text | logprobsparallel tool callspredicted outputsreasoningstructured outputstool calling |
| llama-4-maverick | 1.0M | images, text | text | logprobsstructured outputstool callingvision |
| llama-4-scout | 1.3M | images, text | text | structured outputstool callingvision |
| llama-guard-4-12b | 164K | images, text | text | logprobsreasoningtool callingvision |
| longcat-2.0 | 1.0M | text | text | prompt cachingreasoningtool calling |
| lyria-3-clip | 1.0M | images, text | audio, text | audio outputvision |
| lyria-3-pro | 1.0M | images, text | audio, text | audio outputvision |
| magnum-v4-72b | 33K | text | text | logprobsstructured outputs |
| mercury-2 | 128K | text | text | prompt cachingreasoningstructured outputstool calling |
| mimo-v2.5 | 1.1M | audio, images, text, video | text | audio inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvision |
| mimo-v2.5-pro | 1.1M | text | text | logprobsprompt cachingreasoningstructured outputstool calling |
| minimax-01 | 1.0M | images, text | text | vision |
| minimax-m1 | 1M | text | text | logprobsreasoningtool calling |
| minimax-m2 | 125K | text | text | logprobsreasoningreasoning requiredstructured outputstool calling |
| minimax-m2-her | 66K | text | text | prompt caching |
| minimax-m2.1 | 125K | text | text | logprobsprompt cachingreasoningreasoning requiredtool calling |
| minimax-m2.5 | 125K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool calling |
| minimax-m2.7 | 125K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool calling |
| minimax-m3 | 500K | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvision |
| ministral-14b | 262K | images, text | text | prompt cachingstructured outputstool callingvision |
| ministral-3b | 131K | images, text | text | prompt cachingstructured outputstool callingvision |
| ministral-8b | 128K | images, text | text | parallel tool callspredicted outputsstructured outputstool calling |
| mistral-medium | 120K | images, text | text | parallel tool callspredicted outputstool callingvision |
| mistral-medium-3 | 131K | files, images, text | text | file inputprompt cachingstructured outputstool callingvision |
| mistral-medium-3-5 | 262K | files, images, text | text | file inputreasoningstructured outputstool callingvision |
| mistral-medium-3.1 | 131K | files, images, text | text | file inputprompt cachingstructured outputstool callingvision |
| mistral-nemo | 96K | text | text | logprobsstructured outputstool calling |
| mistral-saba | 33K | files, text | text | file inputprompt cachingstructured outputstool calling |
| mistral-small | 96K | images, text | text | parallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvision |
| mistral-small-24b-instruct | 33K | text | text | structured outputs |
| mistral-small-3.1-24b-instruct | 128K | images, text | text | logprobsvision |
| mistral-small-3.2-24b-instruct | 131K | images, text | text | logprobsstructured outputstool callingvision |
| mixtral-8x22b-instruct | 66K | files, text | text | file inputprompt cachingstructured outputstool calling |
| morph-v3-fast | 82K | text | text | |
| morph-v3-large | 262K | text | text | logprobsstructured outputs |
| muse-glimmer-30b | 131K | images, text | text | logprobsprompt cachingreasoningreasoning requiredstructured outputstool callingvision |
| muse-spark-1.1 | 1.0M | audio, files, images, text, video | text | audio inputfile inputprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| muse-spark-1.2 | 1.0M | audio, files, images, text, video | text | audio inputfile inputprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| muse-spark-1.2-contributor | 1.0M | audio, files, images, text, video | text | audio inputfile inputprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvisionweb search |
| mythomax-l2-13b | 4K | text | text | logprobsreasoningstructured outputstool calling |
| nemotron-3-nano-30b-a3b | 262K | text | text | logprobsprompt cachingreasoningstructured outputstool calling |
| nemotron-3-nano-omni-30b-a3b-reasoning | 256K | audio, images, text, video | text | audio inputreasoningtool callingvideo inputvision |
| nemotron-3-super-120b-a12b | 1M | text | text | logprobsparallel tool callsreasoningstructured outputstool calling |
| nemotron-3-ultra-550b-a55b | 512K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| nemotron-3.5-content-safety | 128K | images, text | text | reasoningvision |
| nemotron-3.5-lightning | 262K | text | text | logprobsparallel tool callsprompt cachingreasoningstructured outputstool calling |
| nemotron-nano-12b-v2-vl | 128K | images, text, video | text | reasoningtool callingvideo inputvision |
| nemotron-nano-9b-v2 | 128K | text | text | reasoningstructured outputstool calling |
| nex-n2-mini | 262K | images, text | text | logprobsprompt cachingreasoningstructured outputstool callingvision |
| nex-n2-pro | 262K | images, text | text | logprobsprompt cachingreasoningtool callingvision |
| nova-2-lite-v1 | 1M | files, images, text, video | text | file inputreasoningtool callingvideo inputvision |
| nova-lite-v1 | 300K | images, text | text | tool callingvision |
| nova-micro-v1 | 128K | text | text | tool calling |
| nova-premier-v1 | 1M | images, text | text | prompt cachingtool callingvision |
| nova-pro-v1 | 300K | images, text | text | tool callingvision |
| o1 | 200K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| o1-pro | 200K | files, images, text | text | file inputreasoningstructured outputsvisionweb search |
| o3 | 200K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| o3-mini | 200K | files, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingweb search |
| o3-mini-high | 200K | files, text | text | file inputprompt cachingreasoningreasoning requiredstructured outputstool callingweb search |
| o3-pro | 200K | files, images, text | text | file inputreasoningstructured outputstool callingvisionweb search |
| o4-mini | 200K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| o4-mini-high | 200K | files, images, text | text | file inputprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| olmo-3-32b-think | 66K | text | text | reasoningreasoning requiredstructured outputs |
| ox-alpha | 1.0M | images, text, video | text | reasoningreasoning requiredtool callingvideo inputvision |
| palmyra-x5 | 1.0M | text | text | |
| perceptron-mk1 | 33K | images, text, video | text | reasoningstructured outputsvideo inputvision |
| phi-4 | 16K | text | text | logprobsreasoningstructured outputstool calling |
| qwen-2.5-72b-instruct | 33K | text | text | logprobsreasoningstructured outputstool calling |
| qwen-2.5-7b-instruct | 33K | text | text | structured outputstool calling |
| qwen-plus | 1M | text | text | logprobsprompt cachingstructured outputstool calling |
| qwen-plus:thinking | 1M | text | text | logprobsreasoningstructured outputstool calling |
| qwen2.5-vl-72b-instruct | 128K | images, text | text | logprobsprompt cachingstructured outputsvision |
| qwen3-14b | 41K | text | text | logprobsreasoningstructured outputstool calling |
| qwen3-235b-a22b | 131K | text | text | logprobsreasoningtool calling |
| qwen3-235b-a22b-thinking | 262K | text | text | logprobsparallel tool callspredicted outputsreasoningreasoning requiredtool calling |
| qwen3-30b-a3b | 41K | text | text | logprobsreasoningtool calling |
| qwen3-30b-a3b-instruct | 262K | text | text | logprobsstructured outputstool calling |
| qwen3-30b-a3b-thinking | 82K | text | text | reasoningreasoning requiredtool calling |
| qwen3-32b | 41K | text | text | logprobsparallel tool callsreasoningstructured outputstool calling |
| qwen3-8b | 131K | text | text | reasoningtool calling |
| qwen3-max | 262K | text | text | logprobsprompt cachingreasoningstructured outputstool calling |
| qwen3-max-thinking | 262K | text | text | logprobsreasoningstructured outputstool calling |
| qwen3-next-80b-a3b-instruct | 262K | text | text | logprobsprompt cachingreasoningstructured outputstool calling |
| qwen3-next-80b-a3b-thinking | 128K | text | text | logprobsparallel tool callsreasoningreasoning requiredstructured outputstool calling |
| qwen3-vl-235b-a22b-instruct | 262K | images, text | text | logprobsprompt cachingreasoningstructured outputstool callingvision |
| qwen3-vl-235b-a22b-thinking | 131K | images, text | text | logprobsreasoningreasoning requiredstructured outputstool callingvision |
| qwen3-vl-30b-a3b-instruct | 262K | images, text | text | logprobsstructured outputstool callingvision |
| qwen3-vl-30b-a3b-thinking | 262K | images, text | text | logprobsreasoningreasoning requiredstructured outputstool callingvision |
| qwen3-vl-32b-instruct | 131K | images, text | text | logprobsstructured outputstool callingvision |
| qwen3-vl-8b-instruct | 262K | images, text | text | logprobsstructured outputstool callingvision |
| qwen3-vl-8b-thinking | 131K | images, text | text | logprobsreasoningreasoning requiredstructured outputstool callingvision |
| qwen3.5-122b-a10b | 262K | images, text, video | text | logprobsreasoningstructured outputstool callingvideo inputvision |
| qwen3.5-27b | 262K | images, text, video | text | logprobsreasoningstructured outputstool callingvideo inputvision |
| qwen3.5-35b-a3b | 262K | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvision |
| qwen3.5-397b-a17b | 230K | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvision |
| qwen3.5-9b | 262K | images, text, video | text | logprobsparallel tool callspredicted outputsreasoningstructured outputstool callingvideo inputvision |
| qwen3.5-flash | 1M | images, text, video | text | reasoningstructured outputstool callingvideo inputvision |
| qwen3.5-plus | 1M | images, text, video | text | logprobsreasoningstructured outputstool callingvideo inputvision |
| qwen3.6-27b | 262K | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvision |
| qwen3.6-35b-a3b | 262K | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvision |
| qwen3.6-flash | 1M | images, text, video | text | logprobsreasoningstructured outputstool callingvideo inputvision |
| qwen3.6-max | 262K | text | text | logprobsreasoningstructured outputstool calling |
| qwen3.6-plus | 1M | images, text, video | text | logprobsparallel tool callspredicted outputsreasoningstructured outputstool callingvideo inputvision |
| qwen3.7-flash | 1M | images, text, video | text | logprobsprompt cachingreasoningtool callingvideo inputvision |
| qwen3.7-max | 1M | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| qwen3.7-plus | 1M | images, text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvision |
| qwen3.8-2.4t-a95b | 1.0M | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool calling |
| qwen3.8-27b | 1M | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvideo inputvision |
| qwen3.8-max | 1M | images, text, video | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvision |
| reka-edge | 16K | images, text, video | text | logprobsstructured outputstool callingvideo inputvision |
| reka-flash-3 | 66K | text | text | logprobsreasoningreasoning requiredstructured outputs |
| relace-apply-3 | 256K | text | text | |
| relace-search | 256K | text | text | tool calling |
| remm-slerp-l2-13b | 6K | text | text | logprobsstructured outputs |
| ring-2.6-1t | 262K | text | text | prompt cachingreasoningreasoning requiredtool calling |
| rocinante-12b | 66K | text | text | logprobsstructured outputs |
| sakana-namazu | 262K | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| seed-1.6 | 262K | images, text, video | text | reasoningstructured outputstool callingvideo inputvision |
| seed-1.6-flash | 262K | images, text, video | text | reasoningstructured outputstool callingvideo inputvision |
| seed-2-1-turbo | 262K | images, text, video | text | parallel tool callspredicted outputsreasoningstructured outputstool callingvideo inputvision |
| seed-2.0-lite | 262K | images, text, video | text | reasoningstructured outputstool callingvideo inputvision |
| seed-2.0-mini | 262K | images, text, video | text | logprobsreasoningstructured outputstool callingvideo inputvision |
| skyfall-36b-v2 | 33K | text | text | logprobsprompt cachingstructured outputs |
| solar-pro-3 | 120K | text | text | prompt cachingreasoningstructured outputstool calling |
| solar-pro-4 | 500K | text | text | prompt cachingreasoningstructured outputstool calling |
| sonar | 120K | images, text | text | reasoningtool callingvisionweb search |
| sonar-deep-research | 128K | text | text | reasoningtool callingweb search |
| sonar-pro | 180K | images, text | text | reasoningtool callingvisionweb search |
| sonar-pro-search | 200K | images, text | text | reasoningreasoning requiredstructured outputsvisionweb search |
| sonar-reasoning-pro | 128K | images, text | text | reasoningtool callingvisionweb search |
| step-3.5-flash | 262K | text | text | reasoningreasoning requiredtool calling |
| step-3.7-flash | 230K | images, text, video | text | logprobsprompt cachingreasoningreasoning requiredstructured outputstool callingvideo inputvision |
| trinity-large-thinking | 262K | text | text | logprobsprompt cachingreasoningreasoning requiredstructured outputstool calling |
| ui-tars-1.5-7b | 128K | images, text | text | logprobsprompt cachingstructured outputsvision |
| unslopnemo-12b | 1.0M | text | text | logprobsstructured outputstool calling |
| virtuoso-large | 131K | text | text | tool calling |
| voxtral-small-24b | 32K | audio, files, text | text | audio inputfile inputprompt cachingstructured outputstool calling |
| weaver | 8K | text | text | logprobs |
| wizardlm-2-8x22b | 66K | text | text | logprobsreasoningtool calling |
Images
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| auto | 2M | audio, files, images, text, video | images, text | audio inputfile inputimage outputlogprobspredicted outputsreasoningstructured outputstool callingvideo inputvisionweb search |
| flux-1-kontext-max | not stated | images, text | images | image editingimage outputvision |
| flux-1-kontext-pro | not stated | images, text | images | image editingimage outputvision |
| flux-1-krea-dev | not stated | text | images | image output |
| flux-1-pro | not stated | text | images | image output |
| flux-1.1-pro | not stated | text | images | image output |
| flux-1.1-pro-ultra | not stated | text | images | image output |
| flux-2-flex | 67K | images, text | images | image editingimage outputvision |
| flux-2-klein-4b | 41K | images, text | images | image editingimage outputvision |
| flux-2-max | 47K | images, text | images | image editingimage outputvision |
| flux-2-pro | 47K | images, text | images | image editingimage outputvision |
| gemini-2.5-flash-image | 30K | images, text | images, text | image outputprompt cachingreasoningstructured outputstool callingvisionweb search |
| gemini-3-pro-image | 60K | audio, files, images, text | images, text | audio inputfile inputimage outputlogprobsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gemini-3.1-flash-image | 60K | audio, files, images, text | images, text | audio inputfile inputimage outputreasoningstructured outputstool callingvisionweb search |
| gemini-3.1-flash-lite-image | 66K | audio, files, images, text | images, text | audio inputfile inputimage outputreasoningtool callingvisionweb search |
| gpt-5-image | 400K | files, images, text | images, text | file inputimage outputlogprobsprompt cachingreasoningreasoning requiredstructured outputsvisionweb search |
| gpt-5-image-mini | 400K | files, images, text | images, text | file inputimage outputlogprobsprompt cachingreasoningreasoning requiredstructured outputsvisionweb search |
| gpt-5.4-image-2 | 272K | files, images, text | images, text | file inputimage outputlogprobsprompt cachingreasoningstructured outputsvisionweb search |
| gpt-image-1 | 400K | images, text | images | image editingimage outputlogprobsprompt cachingstructured outputsvisionweb search |
| gpt-image-1-mini | 400K | images, text | images | image outputlogprobsprompt cachingstructured outputsvisionweb search |
| gpt-image-1.5 | not stated | images, text | images | image editingimage outputvision |
| gpt-image-2 | 32K | images, text | images | image editingimage outputlogprobsprompt cachingstructured outputsvisionweb search |
| grok-imagine-image-2.0 | 66K | images, text | images | image outputlogprobsvision |
| grok-imagine-image-quality | 66K | images, text | images | image outputlogprobsvision |
| hunyuan-image-3 | not stated | text | images | image output |
| hydra-banana | 30K | images, text | images, text | image outputvision |
| hydra-banana-pro | 30K | images, text | images, text | image outputvision |
| imagen-4 | not stated | text | images | image output |
| krea-2-large | 66K | images, text | images | image outputvision |
| krea-2-medium | 66K | images, text | images | image outputvision |
| krea-2-medium-turbo | 66K | images, text | images | image outputvision |
| mai-image-2.5 | 4K | images, text | images | image outputvision |
| mai-image-2.5-pro | 4K | images, text | images | image outputvision |
| qwen-image | not stated | images, text | images | image editingimage outputvision |
| qwen-image-3 | 66K | images, text | images | image outputvision |
| qwen-image-3-pro | 66K | images, text | images | image outputvision |
| qwen-image-edit | not stated | images, text | images | image editingimage outputvision |
| realtime | 2K | text | images | image output |
| recraft-v3 | 66K | images, text | images | image outputvision |
| recraft-v4 | 66K | images, text | images | image outputvision |
| recraft-v4-pro | 66K | images, text | images | image outputvision |
| recraft-v4-pro-vector | 66K | images, text | images | image outputvision |
| recraft-v4-vector | 66K | images, text | images | image outputvision |
| recraft-v4.1 | 66K | images, text | images | image outputvision |
| recraft-v4.1-pro | 66K | images, text | images | image outputvision |
| recraft-v4.1-pro-vector | 66K | images, text | images | image outputvision |
| recraft-v4.1-utility | 66K | images, text | images | image outputvision |
| recraft-v4.1-utility-pro | 66K | images, text | images | image outputvision |
| recraft-v4.1-vector | 66K | images, text | images | image outputvision |
| riverflow-v2-fast | 8K | images, text | images | image outputvision |
| riverflow-v2-pro | 8K | images, text | images | image outputvision |
| riverflow-v2.5-fast | 33K | images, text | images | image outputreasoningreasoning requiredvision |
| riverflow-v2.5-pro | 33K | images, text | images | image outputreasoningreasoning requiredvision |
| seedream-4 | not stated | images, text | images | image editingimage outputvision |
| seedream-4.5 | 4K | images, text | images | image editingimage outputvision |
| seedream-5-0-pro | not stated | images, text | images | image outputvision |
| seedream-5-lite | not stated | images, text | images | image editingimage outputvision |
| z-image-turbo | 2K | text | images | image output |
Coding
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| claude-haiku-4.5 | 180K | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-opus-4 | 200K | files, images, text | text | file inputprompt cachingreasoningtool callingvisionweb search |
| claude-opus-4.1 | 200K | files, images, text | text | file inputprompt cachingreasoningtool callingvisionweb search |
| claude-opus-4.5 | 200K | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-opus-4.6 | 120K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-opus-4.7 | 120K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-opus-4.7-fast | 1M | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-opus-4.8 | 120K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-opus-4.8-fast | 1M | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-opus-5 | 1M | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-opus-5-fast | 1M | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-sonnet-4 | 1M | files, images, text | text | file inputprompt cachingreasoningtool callingvisionweb search |
| claude-sonnet-4.5 | 120K | files, images, text | text | file inputprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-sonnet-4.6 | 120K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| claude-sonnet-5 | 800K | files, images, text | text | file inputlogprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| codestral | 256K | files, text | text | file inputparallel tool callspredicted outputsprompt cachingstructured outputstool calling |
| deepseek-chat | 164K | text | text | logprobsreasoningstructured outputstool calling |
| deepseek-chat-v3 | 164K | text | text | logprobsstructured outputstool calling |
| deepseek-chat-v3.1 | 33K | text | text | logprobsprompt cachingreasoningstructured outputstool calling |
| deepseek-r1 | 64K | text | text | reasoningreasoning requiredstructured outputstool calling |
| deepseek-r1-distill-llama-70b | 8K | text | text | reasoning |
| deepseek-v3.1-terminus | 164K | text | text | prompt cachingreasoningstructured outputstool calling |
| deepseek-v3.2 | 120K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| deepseek-v3.2-exp | 164K | text | text | logprobsreasoningstructured outputstool calling |
| gpt-5-codex | 400K | files, images, text | text | file inputparallel tool callspredicted outputsreasoningtool callingvisionweb search |
| gpt-5.1-codex | 400K | images, text | text | parallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5.1-codex-max | 400K | images, text | text | parallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5.1-codex-mini | 400K | images, text | text | parallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| gpt-5.2-codex | 400K | images, text | text | parallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvisionweb search |
| gpt-5.3-codex | 380K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvisionweb search |
| kat-coder-air-v2.5 | 256K | text | text | logprobsprompt cachingstructured outputstool calling |
| kat-coder-pro-v2 | 262K | text | text | logprobsprompt cachingstructured outputstool calling |
| kat-coder-pro-v2.5 | 256K | text | text | logprobsprompt cachingstructured outputstool calling |
| kimi-k2.7-code | 262K | images, text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningreasoning requiredstructured outputstool callingvision |
| mistral-large | 120K | files, images, text | text | file inputparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool callingvision |
| north-mini-code | 256K | text | text | reasoningtool calling |
| qwen-2.5-coder-32b-instruct | 33K | text | text | |
| qwen3-coder | 262K | text | text | logprobsparallel tool callspredicted outputsprompt cachingreasoningstructured outputstool calling |
| qwen3-coder-30b-a3b-instruct | 262K | text | text | logprobsstructured outputstool calling |
| qwen3-coder-flash | 1M | text | text | logprobsprompt cachingtool calling |
| qwen3-coder-next | 262K | text | text | logprobsprompt cachingreasoningstructured outputstool calling |
| qwen3-coder-plus | 1M | text | text | logprobsprompt cachingreasoningstructured outputstool calling |
| seed-2.0-code | 262K | images, text, video | text | logprobsreasoningstructured outputstool callingvideo inputvision |
Embeddings
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| all-minilm-l12-v2 | 512 | text | embeddings | |
| all-minilm-l6-v2 | 512 | text | embeddings | |
| all-mpnet-base-v2 | 512 | text | embeddings | |
| bge-base-en-v1.5 | 512 | text | embeddings | |
| bge-large-en-v1.5 | 512 | text | embeddings | |
| bge-m3 | 8K | text | embeddings | |
| codestral-embed | 8K | text | embeddings | structured outputs |
| e5-base-v2 | 512 | text | embeddings | |
| e5-large-v2 | 512 | text | embeddings | |
| gemini-embedding-001 | 20K | text | embeddings | |
| gemini-embedding-2 | 8K | audio, files, images, text, video | embeddings | audio inputfile inputvideo inputvision |
| gte-base | 512 | text | embeddings | |
| gte-large | 512 | text | embeddings | |
| lfm-2.5-embedding-350m | 512 | text | embeddings | |
| llama-nemotron-embed-vl-1b-v2 | 131K | images, text | embeddings | vision |
| mistral-embed | 8K | text | embeddings | structured outputs |
| multi-qa-mpnet-base-dot-v1 | 512 | text | embeddings | |
| multilingual-e5-large | 512 | text | embeddings | |
| nemotron-3-embed-1b | 33K | text | embeddings | |
| paraphrase-minilm-l6-v2 | 512 | text | embeddings | |
| pplx-embed-v1-0.6b | 32K | text | embeddings | web search |
| pplx-embed-v1-4b | 32K | text | embeddings | web search |
| qwen3-embedding-4b | 33K | text | embeddings | |
| qwen3-embedding-8b | 33K | text | embeddings | logprobs |
| text-embedding-3-large | 8K | text | embeddings | logprobsstructured outputs |
| text-embedding-3-small | 8K | text | embeddings | logprobsstructured outputs |
| text-embedding-ada | 8K | text | embeddings | |
| text-embedding-ada-002 | 8K | text | embeddings | logprobsstructured outputs |
| voyage-4 | 32K | text | embeddings | |
| voyage-4-large | 32K | text | embeddings | |
| voyage-4-lite | 32K | text | embeddings | |
| voyage-code-4 | 32K | text | embeddings | |
| voyage-multimodal-3.5 | 32K | images, text | embeddings | vision |
Video
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| aleph-2 | not stated | images, text, video | video | video inputvideo outputvision |
| flux-3-video | not stated | images, text, video | video | video inputvideo outputvision |
| flux-video-upscale | not stated | text, video | video | video inputvideo output |
| gen-4.5 | not stated | images, text | video | video outputvision |
| grok-imagine-video | not stated | images, text | video | logprobsvideo outputvision |
| grok-imagine-video-1.5 | not stated | images, text | video | logprobsvideo outputvision |
| hailuo-2.3 | not stated | images, text | video | video outputvision |
| hailuo-3 | not stated | audio, images, text, video | video | audio inputvideo inputvideo outputvision |
| happyhorse-1.0 | not stated | images, text | video | video outputvision |
| happyhorse-1.1 | not stated | images, text | video | video outputvision |
| kling-v3.0-pro | not stated | images, text | video | video outputvision |
| kling-v3.0-std | not stated | images, text | video | video outputvision |
| kling-video-o1 | not stated | images, text | video | video outputvision |
| seedance-1-5-pro | not stated | images, text | video | video outputvision |
| seedance-2.0 | not stated | audio, images, text, video | video | audio inputvideo inputvideo outputvision |
| seedance-2.0-fast | not stated | audio, images, text, video | video | audio inputvideo inputvideo outputvision |
| seedance-2.0-mini | not stated | audio, images, text, video | video | audio inputvideo inputvideo outputvision |
| seedance-2.5 | not stated | audio, images, text, video | video | audio inputvideo inputvideo outputvision |
| sora-2-pro | not stated | images, text | video | logprobsvideo outputvision |
| veo-3.1 | not stated | images, text | video | video outputvision |
| veo-3.1-fast | not stated | images, text | video | video outputvision |
| veo-3.1-lite | not stated | images, text | video | video outputvision |
| wan-2.6 | not stated | images, text | video | video outputvision |
| wan-2.7 | not stated | images, text | video | video outputvision |
Speech to text
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| chirp-3 | not stated | audio | transcription | audio input |
| gpt-4o-mini-transcribe | 128K | audio | transcription | audio inputlogprobsstructured outputs |
| gpt-4o-transcribe | 128K | audio | text, transcription | audio inputlogprobsstructured outputs |
| gpt-transcribe | not stated | audio | transcription | audio inputlogprobsstructured outputs |
| grok-stt-1.0 | not stated | audio | transcription | audio inputlogprobs |
| mai-transcribe-1.5 | not stated | audio | transcription | audio input |
| nemotron-3.5-asr-streaming-multilingual-0.6b | not stated | audio | transcription | audio input |
| nova-3 | not stated | audio | transcription | audio input |
| parakeet-tdt-0.6b-v3 | not stated | audio | transcription | audio input |
| qwen3-asr-0.6b | not stated | audio | transcription | audio input |
| qwen3-asr-1.7b | not stated | audio | transcription | audio input |
| qwen3-asr-flash | not stated | audio | transcription | audio input |
| transcribe-1 | not stated | audio | transcription | audio input |
| voxtral-mini-3b | not stated | audio | transcription | audio input |
| voxtral-mini-transcribe | not stated | audio | transcription | audio inputstructured outputs |
| voxtral-small-24b-stt | not stated | audio | transcription | audio input |
| whisper-1 | not stated | audio | transcription | audio inputlogprobsstructured outputs |
| whisper-large-v3 | not stated | audio | text, transcription | audio inputspeech translation |
| whisper-large-v3-turbo | not stated | audio | text, transcription | audio inputspeech translation |
Text to speech
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| aura-2 | not stated | text | audio | audio output |
| csm-1b | 4K | text | audio | audio output |
| gemini-3.1-flash-tts | 33K | text | audio | audio output |
| grok-voice-tts-1.0 | 15K | text | audio | audio outputlogprobs |
| kokoro-82m | 4K | text | audio | audio output |
| mai-voice-2 | not stated | text | audio | audio output |
| mai-voice-2-flash | not stated | text | audio | audio output |
| orpheus-3b-0.1-ft | 4K | text | audio | audio output |
| qwen-audio-3.0-tts-flash | not stated | text | audio | audio output |
| qwen-audio-3.0-tts-plus | not stated | text | audio | audio output |
| s1 | not stated | text | audio | audio output |
| s2-pro | not stated | text | audio | audio output |
| s2.1-pro | not stated | text | audio | audio output |
| speech-2.8-hd | not stated | text | audio | audio output |
| speech-2.8-turbo | not stated | text | audio | audio output |
| voxtral-mini-tts | 4K | text | audio | audio outputstructured outputs |
Reranking
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| llama-nemotron-rerank-vl-1b-v2 | 10K | images, text | rerank | vision |
| qwen3-reranker-8b | 41K | text | rerank | logprobsstructured outputs |
| rerank-2.5 | 32K | text | rerank | |
| rerank-2.5-lite | 32K | text | rerank | |
| rerank-4-fast | 33K | text | rerank | structured outputs |
| rerank-4-pro | 33K | text | rerank | structured outputs |
| rerank-v3.5 | 4K | text | rerank | structured outputs |
Moderation
| Model | Context | Input | Output | Capabilities |
|---|---|---|---|---|
| mistral-moderation | not stated | images, text | moderation | vision |
| omni-moderation | not stated | images, text | moderation | vision |
Context is the smallest window across the providers serving the model.