Skip to Content
ReferenceModels

Model Catalog

v0.7.0-alpha snapshot. Model ids move fast. This page tracks the curated ids shipped in the setup wizard as of the current release. The live source of truth is curated_models_for_provider / default_model_for_provider in src/onboard/wizard.rs; the /model picker in the TUI reads the same lists.

Model ids are free-form strings — the runtime passes whatever you set in default_model straight to the provider. Nothing gates them, so any id the provider accepts works even if it is not listed here. The curated ids below are just the vetted defaults the onboarding wizard and /model picker offer.

How the per-provider default works

When you run the setup wizard and accept the suggested model for a provider, the wizard writes that provider’s default into default_model (see default_model_for_provider in src/onboard/wizard.rs). This is a wizard-time default — the id it pre-fills for the chosen provider — not a live runtime fallback.

If default_model is left unset entirely in config.toml, the agent loop uses a hardcoded fallback of anthropic/claude-sonnet-4 rather than the per-provider default (see src/agent/loop_.rs). In practice default_model is always populated after onboarding, so the per-provider default is what you get.

Canonical provider name (aliases resolve to these) and the id default_model_for_provider returns:

ProviderDefault model id
openrouteranthropic/claude-sonnet-4.6
anthropicclaude-sonnet-4-6
openaigpt-5.5
openai-codexgpt-5.3-codex
gemini (google)gemini-3-pro
deepseekdeepseek-v4-pro
xai (grok)grok-4-1-fast-reasoning
qwenqwen3.7-plus
qwen-codeqwen3.6-coder-plus
moonshot (kimi)kimi-k2.6
kimi-codekimi-for-coding
glm / zaiglm-5.2
minimaxMiniMax-M3
mistralmistral-large-latest
coherecommand-a-03-2025
perplexitysonar-pro
groqllama-spark
venicezai-org-glm-5.1
fireworksaccounts/fireworks/models/llama-spark
together-aimeta-llama/Llama-Spark
nvidiameta/llama-spark
ollamallama-spark
llamacppggml-org/llama-spark-GGUF
bedrockanthropic.claude-sonnet-4-6-v1:0

Curated menus (flagships)

The wizard and /model picker present these per-provider menus. The entry marked “(default)” is what default_model_for_provider returns — note that Moonshot and Qwen list the newest id first while still defaulting to the stable flagship.

Anthropic

Model idNotes
claude-sonnet-4-6Balanced, recommended (default)
claude-opus-4-8Best quality
claude-opus-4-7Previous flagship
claude-haiku-4-5-20251001Fastest, cheapest
claude-sonnet-4-5-20250929Legacy

OpenAI

Model idNotes
gpt-5.5Latest flagship, recommended (default)
gpt-5.4Previous flagship, stable
gpt-5.3-codexAgentic coding
gpt-5.2Older, stable
gpt-5-miniFaster, cheaper
gpt-5-nanoLowest latency/cost
o4-miniReasoning fallback

Removed phantom ids (0.6.80). The openai list no longer includes gpt-5.5-codex, and the bare gpt-5.3 id is gone — neither exists on the OpenAI API. The openai-codex provider now defaults to gpt-5.3-codex (the newest codex id that actually exists), and regression tests assert the phantom ids never reappear.

OpenAI Codex (openai-codex)

Model idNotes
gpt-5.3-codexLatest, recommended (default)
gpt-5.2-codexOlder flagship
gpt-5.1-codex-miniCheap, capable
gpt-5-codexLegacy, stable
o4-miniReasoning fallback

DeepSeek

Model idNotes
deepseek-v4-proLatest flagship, recommended (default)
deepseek-v4-flashFast, cost-efficient
deepseek-chatMapped to V3.2 non-thinking
deepseek-reasonerMapped to V3.2 thinking

Qwen

Model idNotes
qwen3.7-maxLatest flagship, highest quality
qwen3.7-plusBalanced (default)
qwen3.6-maxPrevious flagship
qwen3.6-plusBalanced
qwen3.6-turboFast, cost-efficient

GLM / Z.ai (glm, zai)

Model idNotes
glm-5.2Latest flagship, recommended (default)
glm-5.1Previous flagship
glm-5Stable
glm-4.7Strong general-purpose, stable
glm-4.5-airLower latency

Moonshot / Kimi (moonshot)

Model idNotes
kimi-k2.7-codeLatest, agentic coding
kimi-k2.6Flagship, recommended (default)
kimi-k2.5Previous flagship
kimi-k2-thinkingDeep reasoning + tool use
kimi-k2-0905-previewLegacy

MiniMax

Model idNotes
MiniMax-M3Latest flagship, recommended (default)
MiniMax-M2.7Previous flagship
MiniMax-M2.7-highspeedFast
MiniMax-M2.5Stable
MiniMax-M2.1Strong coding/reasoning, stable

xAI / Grok (xai)

Model idNotes
grok-4-1-fast-reasoningRecommended (default)
grok-4-1-fast-non-reasoningLow latency
grok-code-fast-1Coding specialist
grok-4Max quality

Mistral

Model idNotes
mistral-large-latestLatest flagship (default)
mistral-medium-latestBalanced
codestral-latestCode-focused
devstral-latestSoftware-engineering specialist

Cohere

Model idNotes
command-a-03-2025Flagship enterprise model (default)
command-a-reasoning-08-2025Agentic reasoning
command-r-08-2024Stable fast baseline

Gemini

Model idNotes
gemini-3-proLatest frontier reasoning, recommended (default)
gemini-3-flashBest price/performance
gemini-2.5-proPrevious flagship, stable
gemini-2.5-flashLegacy fast tier
gemini-2.5-flash-liteLowest cost

OpenRouter and aggregators

openrouter (and astrai) use slash-prefixed, cross-provider ids because they route to many upstreams. The curated OpenRouter menu leads with anthropic/claude-sonnet-4.6 (the default), and also offers anthropic/claude-opus-4.7, openai/gpt-5.5, google/gemini-3-pro, x-ai/grok-4.1-fast, deepseek/deepseek-v3.2, moonshotai/kimi-k2.6, and z-ai/glm-5.1, among others. Set any provider-prefixed id the aggregator supports.

Overriding at runtime

For most providers the wizard can also fetch the live model list from the provider’s /models endpoint (supports_live_model_fetch in src/onboard/wizard.rs), so the picker shows current ids beyond this curated set. You can always set default_model by hand or switch mid-session with /model. See Providers for provider slugs, routing, and fallbacks.

Last updated on