Models

Every LLM Prism can route to. We're a control plane, not a marketplace — every provider is a direct integration, and every model has a specific routing role (eco / balanced / sport, or specialty fallback). No middleman markup; no marketplace fees baked in.

Providers
8
7 active
Models
27
curated, routing-table-aware
Architectures
8
Claude / GPT / Gemini / Llama / Qwen / DeepSeek / Mistral / Grok / GLM / Kimi
Last updated
2026-05-22
catalog snapshot

Providers

All direct integrations. Excluded means registered but not currently routed to — e.g. DeepSeek is excluded until a customer asks for it, at which point we fund the account and flip it on.

ProviderArchitecture familyModelsStatus
AnthropicClaude5Active
CerebrasLlama/Qwen2Active
DeepSeekDeepSeek2Excluded
FireworksKimi/GLM2Active
GoogleGemini3Active
GroqLlama/Qwen/GPT-OSS5Active
MistralMistral5Active
OpenAIGPT3Active

Models

Cost is per 1M tokens, before Prism markup. The "routes" column shows where each model is picked by mode-based auto-routing today.

ModelProviderCapabilityInput $/1MOutput $/1MAuto-router uses for
cerebras-llama-8bCerebrasSmall (fast)$0.10$0.10code/eco
cerebras-qwen-235bCerebrasFrontier$0.60$1.20code/balanced
claude-haikuAnthropicSmall (fast)$0.80$4.00
claude-opusAnthropicFrontier$15.00$75.00simple/sport, reasoning/sport
claude-opus-4-7AnthropicFrontier$15.00$75.00
claude-sonnetAnthropicLarge$3.00$15.00
claude-sonnet-4-7AnthropicLarge$3.00$15.00
codestralMistralCode-specialized$0.30$0.90
deepseek-v4-flashDeepSeekSmall (fast)$0.07$0.28
deepseek-v4-proDeepSeekFrontier$0.14$0.56
fireworks-glm-5p1FireworksLarge$0.50$2.00
fireworks-kimi-k2FireworksLong-context$0.60$2.50
gemini-3-5-proGoogleLarge$1.25$10.00
gemini-flashGoogleSmall (fast)$0.07$0.30
gemini-proGoogleLarge$1.25$10.00complex/sport
gpt-4oOpenAILarge$2.50$10.00complex/balanced
gpt-4o-miniOpenAISmall (fast)$0.15$0.60
gpt-5-5OpenAIFrontier$2.50$10.00
groq-gpt-ossGroqLarge$0.15$0.75
groq-llama-70bGroqLarge$0.59$0.79complex/eco
groq-llama-8bGroqSmall (fast)$0.05$0.08simple/eco, simple/balanced, reasoning/eco
groq-llama4-scoutGroqLarge$0.11$0.34
groq-qwen-32bGroqMedium$0.29$0.59reasoning/balanced
magistral-mediumMistralReasoning$2.00$5.00
ministral-8bMistralSmall (fast)$0.10$0.10
mistral-largeMistralFrontier$2.00$6.00
mistral-mediumMistralMedium$0.40$2.00code/sport

Try it

Specify a mode (eco / balanced / sport) and Prism picks the right model per request. Override with X-Prism-Model-Prefer to force a specific model (Pro+ for non-incumbent providers).