Seshat AIDocumentation

Documentation / Reference

Providers and models

Every provider Seshat supports, with the models it knows, their context window and their capabilities.

This page is generated from the provider registry of the runtime. How to set keys and endpoints is in Configuration, and the code side in Providers and authentication.

Providers

Provider IDServiceAuthenticationVisionPrompt cache
anthropicAnthropicapi_keyYesYes
openaiOpenAIapi_keyYesYes
mistralMistral AIapi_keyYesNo
geminiGoogle Geminiapi_keyYesNo
deepseekDeepSeekapi_keyYesNo
ollamaOllamanoneNoNo
openrouterOpenRouterapi_keyYesNo
codexCodexoauthNoNo
z-aiZ.aiapi_keyYesYes
minimaxMiniMaxapi_keyYesYes
opencodeOpenCode Zenapi_keyYesNo
kimiKimi (Moonshot AI)api_keyYesYes

Models

Anthropic (anthropic)

Model IDContextMax outputPrompt cache
claude-sonnet-5-51M128KYes
claude-haiku-5-51M128KYes

Default large and small model shown. 14 models in all, read from the models.dev catalog.

OpenAI (openai)

Model IDContextMax outputPrompt cache
gpt-6.1-sol1M128KYes
gpt-6-luna1M128KYes

Default large and small model shown. 24 models in all, read from the models.dev catalog.

Mistral AI (mistral)

Model IDContextMax outputPrompt cache
mistral-large-2512262K8KNo
mistral-small-2603262K256KNo

Default large and small model shown. 17 models in all, read from the models.dev catalog.

Google Gemini (gemini)

Model IDContextMax outputPrompt cache
gemini-4-argon1M32KNo
gemini-3.5-flash-lite1M65KNo

Default large and small model shown. 18 models in all, read from the models.dev catalog.

DeepSeek (deepseek)

Model IDContextMax outputPrompt cache
deepseek-flash1M393KNo
deepseek-v4-pro1M393KNo

OpenRouter (openrouter)

Model IDContextMax outputPrompt cache
anthropic/claude-sonnet-5.51M128KNo
anthropic/claude-haiku-5.51M128KNo

Default large and small model shown. 311 models in all, read from the models.dev catalog.

Codex (codex)

Model IDContextMax outputPrompt cache
gpt-5.6-sol200K100KNo
gpt-5.6-terra200K100KNo
gpt-5.5200K100KNo
gpt-5.4272K32KNo
gpt-5.6-luna200K100KNo
gpt-5.4-mini272K16KNo

Z.ai (z-ai)

Model IDContextMax outputPrompt cache
glm-5.31M131KYes
glm-5.3-flashx1M131KYes

Default large and small model shown. 18 models in all, read from the models.dev catalog.

MiniMax (minimax)

Model IDContextMax outputPrompt cache
MiniMax-M2.7-highspeed204K131KYes
MiniMax-M31M512KYes
MiniMax-M2.7204K131KYes
MiniMax-M2.5-highspeed204K131KYes
MiniMax-M2.5204K131KYes
MiniMax-M2.1204K131KYes
MiniMax-M2204K131KYes
MiniMax-M3.1-Flash1M512KYes

OpenCode Zen (opencode)

Model IDContextMax outputPrompt cache
claude-sonnet-5-51M128KNo
claude-haiku-5-51M128KNo

Default large and small model shown. 84 models in all, read from the models.dev catalog.

Kimi (Moonshot AI) (kimi)

Model IDContextMax outputPrompt cache
kimi-k31M8KYes
kimi-k2.7-code-highspeed262K8KYes
kimi-k2.6262K8KYes
kimi-k2.7-code262K8KYes

Updated on 2026-10-19