Model profiles¶
Generated from the architecture model
This page is rendered from model/**/*.yaml by tools/docs_model during mkdocs build. Edit the YAML, not this page; hand-written design lives on the linked service, platform and plane pages.
| Profile | Capabilities | Privacy | Providers | Fallback | Context window | Status |
|---|---|---|---|---|---|---|
Code fast (code-fast) |
tool-calling, strict-json, file-editing | Regional | azure-openai:approved small code deployment, anthropic:approved fast Claude deployment, ollama:self-hosted code model | code-agent | 128k | target |
Code agent (code-agent) |
tool-calling, parallel-tool-calling, strict-json, json-schema, file-editing, shell-tool-usage, multi-turn, patch-success | Regional | anthropic:approved Claude coding deployment, azure-openai:approved coding deployment, ollama:self-hosted coding model | code-agent-strong | 128k | target |
Code agent strong (code-agent-strong) |
tool-calling, parallel-tool-calling, strict-json, json-schema, file-editing, shell-tool-usage, multi-turn, patch-success, planning, error-recovery, reliable-context-64k | Regional | anthropic:approved strongest Claude deployment, azure-openai:approved strongest deployment | reasoning-high | 200k | target |
Reasoning (reasoning) |
instruction-following, planning, strict-json, json-schema | Regional | anthropic:approved Claude reasoning deployment, azure-openai:approved reasoning deployment | reasoning-high | 128k | target |
Reasoning high (reasoning-high) |
instruction-following, planning, error-recovery, strict-json, json-schema, reliable-context-64k | Regional | anthropic:approved strongest Claude deployment, azure-openai:approved strongest reasoning deployment | — | 200k | target |
Structured output (structured) |
strict-json, json-schema, instruction-following | Regional | azure-openai:approved structured-output deployment, anthropic:approved Claude deployment with structured outputs | code-agent | 64k | target |
Vision (vision) |
vision, instruction-following, strict-json | Regional | azure-openai:approved multimodal deployment, anthropic:approved multimodal Claude deployment | — | 128k | target |
Long context (long-context) |
reliable-context-64k, instruction-following, multi-turn | Regional | azure-openai:approved long-context deployment, anthropic:approved long-context Claude deployment | reasoning-high | 1m | target |
Embedding (embedding) |
embedding | Regional | azure-openai:approved embedding deployment, ollama:self-hosted embedding model | — | 8k | target |
Reranker (reranker) |
rerank | Regional | azure-ai-inference:approved reranker deployment, ollama:self-hosted reranker | — | 8k | target |
PII local (pii-local) |
tool-calling, strict-json, json-schema, instruction-following | TenantPrivate | ollama:self-hosted model on a tenant-private pool | — | 64k | target |