Skip to main content

Catalog Reference

Ante resolves every provider and model from a catalog: the built-in presets documented here, with your ~/.ante/catalog.json merged on top at startup. The catalog is read once per process — restart Ante to pick up edits.

This page is the exhaustive reference. For how to pick a provider day to day and copy-paste customization recipes, see Providers.

Built-in providers​

ProviderkeyWire styleAuthentication
AnthropicanthropicAnthropicMessageANTHROPIC_API_KEY
Anthropic Subscriptionanthropic-subscriptionAnthropicMessageOAuth
OpenAIopenaiOpenAiResponseOPENAI_API_KEY
OpenAI Compatibleopenai-compatibleOpenAiCompatibleOPENAI_COMPATIBLE_API_KEY
OpenAI Subscriptionopenai-subscriptionOpenAiResponseOAuth (Codex)
GeminigeminiGeminiGEMINI_API_KEY
Vertex AI Geminivertex-geminiGeminiVERTEX_GEMINI_API_KEY
Open RouteropenrouterOpenAiCompatibleOPENROUTER_API_KEY
Open Router Responsesopenrouter-responsesOpenAiResponseOPENROUTER_API_KEY
Open Router Anthropicopenrouter-anthropicAnthropicMessageOPENROUTER_API_KEY
Grok (xAI)xaiOpenAiResponseXAI_API_KEY
ZaizaiOpenAiCompatibleZAI_API_KEY
DeepSeekdeepseekOpenAiCompatibleDEEPSEEK_API_KEY
Ali Coding Planali-coding-planAnthropicMessageALI_CODING_PLAN_API_KEY
AntixantixAnthropicMessageOAuth
Antix (API key)antix-api-keyAnthropicMessageANTIX_API_KEY
LocallocalOpenAiCompatiblenone

The key column is what you pass to --provider. Providers authenticating via OAuth (anthropic-subscription, openai-subscription, antix) sign in interactively through the TUI; the rest read an API key from the listed environment variable.

Models by provider​

Each provider ships a set of preferred_models — the ids that appear in the model picker. When you select a provider without naming a model, Ante defaults to its heaviest-weight model (see Weight classes), not simply the first in the list. Explicit provider choices are authoritative: if you pass --provider, Ante sends the model id you supplied to that provider even when it is not listed in preferred_models (see Resolution).

Anthropic​

anthropic and anthropic-subscription serve the same models:

ModelDescriptionContext
claude-haiku-4-5Fastest model with near-frontier intelligence200K
claude-sonnet-5Best combination of speed and intelligence1M
claude-opus-5Complex agentic coding and enterprise work1M
claude-fable-5-1Highest capability for demanding reasoning1M
ante --provider anthropic --model claude-sonnet-5

OpenAI​

ModelDescriptionContext
gpt-5.6-solGPT-5.6 flagship for complex reasoning and coding1.05M
gpt-5.6-terraGPT-5.6 balance of intelligence and cost1.05M
gpt-5.6-lunaGPT-5.6 model for efficient, high-volume workloads1.05M
gpt-5.5-proHighest reliability on complex problems400K
gpt-5.4-proHighest reliability in the GPT-5.4 family400K
gpt-5.3-codexFrontier coding model400K
gpt-5.4-miniFaster, cheaper GPT-5.4 model400K
gpt-5.4-nanoSmallest GPT-5.4 model400K
ante --provider openai --model gpt-5.6-sol

The openai-subscription provider (ChatGPT/Codex OAuth) serves gpt-5.6-sol, gpt-5.6-terra, and gpt-5.6-luna with 272K context, plus gpt-5.5, gpt-5.4, and gpt-5.4-mini with 400K context.

OpenAI transports​

The openai and openai-subscription providers can stream responses two ways:

TransportEndpointDefault
http_sse (HTTP + SSE)https://api.openai.com/v1/responses (or the subscription equivalent)✅
websocket_auto (persistent WebSocket)wss://.../v1/responsesopt-in

WebSocket mode keeps one connection open per session and continues each turn with previous_response_id plus only the new input items. In long agentic tool loops this eliminates per-turn round-trip latency. See OpenAI's WebSocket mode guide for the underlying protocol.

Opt in via the process-scoped env var (default is http_sse):

ANTE_OPENAI_TRANSPORT=websocket_auto ante --provider openai --model gpt-5.4-mini

Accepted values: http_sse, websocket_auto. Unknown values warn and fall back to http_sse.

Notes:

  • WS only engages on streaming turns (TUI sessions). Headless --prompt mode uses non-streaming chat() and bypasses WS by design.

  • Any transient WS failure (connect timeout, network blip) falls back to HTTP/SSE for that turn — users never see hard errors caused by the transport choice. Auth failures (401/403) flip the entire session to HTTP/SSE for the rest of its lifetime.

  • Works against both api.openai.com (API-key) and the ChatGPT subscription backend at chatgpt.com/backend-api/codex.

  • To verify which transport is active: in debug builds, toggle the debug panel (Ctrl+L) and check the Info tab for the OpenAI Transport: line. Otherwise grep your session log for transport selected:

    grep "transport selected" ~/.ante/logs/*/ante.*.log

Gemini​

gemini and vertex-gemini serve the same models:

ModelDescriptionContext
gemini-3.8-flashGemini 3.8 Flash for long-horizon coding and autonomous agents1.05M
gemini-3.7-flashGemini 3.7 Flash for fast agentic and multimodal workflows1.05M
gemini-3.6-flashGemini 3.6 Flash for agentic and multimodal workflows1M
gemini-3.1-pro-previewGemini 3.1 Pro with deep thinking1M
gemini-3.1-pro-preview-customtoolsGemini 3.1 Pro optimized for custom tool usage1M
gemini-3.5-flash-liteFastest, lowest-cost Gemini 3.5 model for high-throughput work1M
ante --provider gemini --model gemini-3.1-pro-preview

Grok (xAI)​

ModelDescriptionContext
grok-4.6Grok 4.6 flagship reasoning model500K
ante --provider xai --model grok-4.6

Zai​

ModelDescriptionContext
glm-5.3Latest GLM-5.3 model for coding and agentic tasks1M
glm-5.3-flashFast, cost-efficient GLM-5.3 multimodal model1M
ante --provider zai --model glm-5.3

Open Router​

Open Router ships as three providers sharing one API key, split by wire style: openrouter (OpenAI-compatible chat), openrouter-responses (OpenAI Responses), and openrouter-anthropic (Anthropic Messages).

openrouter:

ModelDescriptionContext
deepseek/deepseek-v4-flash-0731DeepSeek V4 Flash 07311.05M
google/gemini-3.8-flashGemini 3.8 Flash for long-horizon coding and autonomous agents1.05M
google/gemini-3.7-flashGemini 3.7 Flash for fast multimodal agentic workflows1.05M
x-ai/grok-4.6Grok 4.6 flagship reasoning model via OpenRouter500K
qwen/qwen3.8-maxQwen 3.8 flagship multimodal reasoning model1M
qwen/qwen3.7-plusCost-effective Qwen 3.7 model with vision1M
qwen/qwen3.7-flashLow-cost Qwen 3.7 multimodal reasoning model1M
meta/muse-spark-1.3Muse Spark 1.3 multimodal agentic reasoning model1.05M
tencent/hy3Tencent Hy3 agentic reasoning model262K
xiaomi/mimo-v2.5-proMiMo-V2.5-Pro flagship agentic model1.05M
xiaomi/mimo-v2.5MiMo-V2.5 omnimodal model with vision1.05M
z-ai/glm-5.3GLM 5.3 coding and agentic reasoning model via OpenRouter1.05M
z-ai/glm-5.3-flashFast, cost-efficient GLM 5.3 multimodal model via OpenRouter1.05M
moonshotai/kimi-k3Kimi K3 flagship multimodal reasoning model1.05M
moonshotai/kimi-k2.7-codeKimi K2.7 Code reasoning model with vision262K
minimax/minimax-m3MiniMax M3 multimodal model512K
deepseek/deepseek-v4-pro-0813DeepSeek V4 Pro 0813 GA1.05M

openrouter-responses serves openai/gpt-5.6-sol, openai/gpt-5.6-sol-pro, openai/gpt-5.6-terra, openai/gpt-5.6-terra-pro, openai/gpt-5.6-luna, and openai/gpt-5.6-luna-pro (1.05M context), plus openai/gpt-5.3-codex (400K). The -pro ids select OpenRouter's pro reasoning route.

openrouter-anthropic serves the Claude family over Anthropic Messages: anthropic/claude-haiku-4.5 (200K context), plus anthropic/claude-sonnet-5, anthropic/claude-opus-5, anthropic/claude-opus-5-fast, and anthropic/claude-fable-5.1 (1M).

Each provider also accepts any model from Open Router's full model list by id, as long as it speaks that provider's wire style.

ante --provider openrouter --model deepseek/deepseek-v4-pro-0813
ante --provider openrouter-responses --model openai/gpt-5.6-sol
ante --provider openrouter-anthropic --model anthropic/claude-fable-5.1

DeepSeek​

ModelDescriptionContext
deepseek-v4-flashDeepSeek V4 Flash1M
deepseek-v4-proDeepSeek V4 Pro1M
ante --provider deepseek --model deepseek-v4-flash

Set DEEPSEEK_API_KEY, and optionally DEEPSEEK_BASE_URL.

Ali Coding Plan​

ModelDescriptionContext
qwen3.5-plusQwen 3.5 Plus with vision1M
qwen3.6-plusLatest Qwen 3.6 Plus with vision1M
glm-5.1GLM-5.1 model200K
MiniMax-M2.5MiniMax M2.5 model205K
qwen3-max-2026-01-23Qwen 3 Max snapshot1M
qwen3-coder-nextQwen 3 Coder Next coding model1M
qwen3-coder-plusQwen 3 Coder Plus coding model1M
qwen-plusFast and capable1M
ante --provider ali-coding-plan --model qwen3.5-plus

Requires a Coding Plan subscription and API key (format: sk-sp-xxxxx). The built-in provider speaks the Anthropic wire endpoint. Some models are only available on the OpenAI-compatible endpoint — add a custom catalog entry to reach it:

{
  "providers": {
    "ali-coding-plan-oai": {
      "display_name": "Ali Coding Plan (OpenAI)",
      "base_url": "https://coding.dashscope.aliyuncs.com/v1",
      "auth": { "bearer": { "env_key": "ALI_CODING_PLAN_API_KEY" } },
      "wire_style": "OpenAiCompatible",
      "http_headers": {
        "User-Agent": "ante-cli/1.0.0",
        "X-DashScope-UserAgent": "coding-agent"
      }
    }
  }
}

Antix​

antix (OAuth) and antix-api-key (API key) serve the same models through Antigma:

ModelDescriptionContext
gemini-3.7-flashGemini 3.7 Flash for fast agentic and multimodal workflows1M
grok-4.5Grok 4.5, xAI flagship for code and agentic work500K
grok-4.6Grok 4.6 flagship reasoning model500K
kimi-k3Kimi K3 flagship reasoning model with vision1M
qwen3.7-maxLatest Qwen 3.7 Max1M
qwen3.7-plusLatest Qwen 3.7 Plus1M
gemini-3.1-pro-previewGemini 3.1 Pro with deep thinking1M
claude-haiku-4-5Fast and capable200K
claude-sonnet-5Best combination of speed and intelligence1M
claude-opus-5Complex agentic coding and enterprise work1M
claude-fable-5-1Most capable Claude model1M
gpt-5.3-codexCoding-optimized Codex model—
gpt-5.4-miniFaster, cheaper GPT-5.4 model400K
gpt-5.4-nanoSmallest GPT-5.4 model400K
deepseek-v4-flashFast DeepSeek V4 Flash1M
deepseek-v4-proDeepSeek V4 Pro with deep thinking1M
glm-5.3Latest GLM-5.3 model for coding and agentic tasks1M
glm-5.3-flashFast GLM-5.3 Flash for coding and agentic tasks1M
ante --provider antix --model claude-sonnet-5         # OAuth
ante --provider antix-api-key --model claude-sonnet-5 # ANTIX_API_KEY

Local​

Run GGUF models locally via the built-in llama.cpp engine — no API key required. local is a wildcard provider; the available model depends on what the offline server has loaded, and the entry exists in the catalog only while a live server is registered. Selecting local with no server running gates the session with a notice until you start or attach one. See Offline Mode for setup.

ante --provider local

catalog.json schema​

~/.ante/catalog.json is merged on top of the built-in presets at startup. It has two optional top-level keys, both maps keyed by id:

{
  "providers": { /* add or override providers */ },
  "models":    { /* override fields on existing models */ }
}

A providers entry whose key matches a built-in (e.g. openai) patches that preset field by field; a new key adds a provider. The map key is the value you pass to --provider. An id (or legacy name) field inside an entry is ignored.

note

~/.ante can be relocated with the ANTE_HOME environment variable; the catalog is always $ANTE_HOME/catalog.json.

Provider fields​

FieldRequiredDescription
display_nameNew providers: optionalFriendly name shown in the TUI. New providers default to the provider map key when omitted.
base_urlNew providers: ✅API base URL.
wire_styleNew providers: ✅API dialect Ante speaks — AnthropicMessage, OpenAiCompatible, OpenAiResponse, or Gemini.
auth—How to authenticate (see Authentication). Omit for a provider that needs no key.
http_headers—Extra headers sent with every request, as a string→string map.
extra_body—Provider-specific JSON fields merged into every chat and streaming request body (see Extra request fields).
supports_web_search—Whether the provider exposes a native web-search affordance. When true, the WebSearch tool is added to sessions on this provider automatically; when false (the default), WebSearch is not offered.
preferred_models—Models shown for this provider in the picker and used as provider defaults (see Model fields). For built-in provider overlays, matching model ids are patched and new ids are appended; omitted leaves the preset list unchanged.
stream_idle_timeout_secs—How long a provider request may remain idle before it fails, in seconds. For streaming calls, this covers both the wait for the first event and silence between later events over SSE or WebSocket; for buffered calls, it bounds the whole request while the response is collected. Defaults to 300. Raise it for a slow local or proxied endpoint; lower it to fail over from a stalled one sooner.
{
  "providers": {
    "my-provider": {
      "display_name": "My Provider",
      "base_url": "https://api.example.com/v1",
      "wire_style": "OpenAiCompatible",
      "auth": { "bearer": { "env_key": "MY_PROVIDER_API_KEY" } },
      "http_headers": { "X-Org": "my-team" },
      "preferred_models": [
        {
          "id": "my-model",
          "description": "My fine-tuned model",
          "max_tokens": 32000,
          "context_limit": 200000,
          "effort": "medium",
          "supported_efforts": ["low", "medium", "high"]
        }
      ]
    }
  }
}

For built-in providers, include only the fields you want to change. Optional fields with null clear the preset value, which is useful when repointing a provider at a local proxy that should not send the built-in auth or headers:

{
  "providers": {
    "openai": {
      "base_url": "http://localhost:8080/v1",
      "auth": null,
      "http_headers": null
    }
  }
}

Extra request fields​

Some gateways and proxies require parameters Ante doesn't model — a service tier, a routing hint, a vendor-specific beta block. extra_body is a JSON object merged into the top level of every chat and streaming request body Ante sends to that provider:

{
  "providers": {
    "my-gateway": {
      "base_url": "https://gateway.example.com/v1",
      "wire_style": "OpenAiCompatible",
      "auth": { "bearer": { "env_key": "MY_GATEWAY_API_KEY" } },
      "extra_body": {
        "service_tier": "priority",
        "routing": { "region": "us-east" }
      }
    }
  }
}

Values are passed through verbatim, so any JSON shape works. On a built-in provider overlay, extra_body merges key by key into the preset's — matching keys are replaced, others are left alone.

extra_body can only add fields, never override the ones Ante generates for the selected wire_style:

wire_styleReserved keys
AnthropicMessagemodel, max_tokens, messages, system, temperature, tools, stream, thinking, stop_sequences, output_config, cache_control
OpenAiCompatiblemodel, messages, reasoning_effort, max_tokens, stream, stream_options, temperature, top_p, top_k, stop, tools, tool_choice, thinking, enable_thinking, thinking_budget, enable_search, search_options
OpenAiResponsemodel, instructions, input, tools, tool_choice, parallel_tool_calls, max_output_tokens, temperature, reasoning, stream, include, store
Geminicontents, systemInstruction, generationConfig, tools, toolConfig

A reserved key is dropped at catalog load and named in the startup notice; the rest of the provider still loads. To change one of these, use the model fields or the environment overrides instead:

Skipped 1 entry in ~/.ante/catalog.json:
  - providers.my-gateway.extra_body: ignoring Ante-owned request field(s): model

Authentication​

auth is a tagged object: the tag chooses where the credential goes, and the credential is either an environment variable (env_key) or one of Ante's built-in OAuth flows (oauth_preset).

FormJSONSends
Bearer token{ "bearer": { "env_key": "MY_KEY" } }Authorization: Bearer <token>
Custom header{ "header": { "name": "x-api-key", "env_key": "MY_KEY" } }<name>: <token>
Query param{ "query": { "name": "api_key", "env_key": "MY_KEY" } }?<name>=<token>

Swap env_key for oauth_preset to authenticate through a built-in OAuth flow (anthropic, openai, or antix) instead of an env var:

"auth": { "bearer": { "oauth_preset": "anthropic" } }

A provider counts as authenticated when its credential is present — the env var is set and non-empty, or the OAuth token exists on disk. Providers with no auth block are always authenticated. Unauthenticated providers are skipped during auto-detection and shown last in the picker.

Model fields​

The same fields describe a model fully inside a provider's preferred_models (where id is required) or partially under models (where the map key supplies the id). Every field except the id is optional:

FieldDescription
idModel id. Required in preferred_models (name is accepted as a legacy alias); supplied by the map key under models.
display_nameFriendly name.
descriptionShort description shown in the model picker.
temperatureSampling temperature. On AnthropicMessage providers, extended thinking requires 1 — any other value with an effort level set fails the turn with an explicit message.
top_pNucleus sampling cutoff.
top_kTop-k sampling cutoff.
max_tokensMaximum output tokens per turn. On AnthropicMessage providers that take an explicit thinking budget, it must exceed 1024 — the API's minimum budget — when an effort level is set.
stop_sequencesList of strings that end generation.
context_limitContext window size, in tokens.
effortDefault effort level — min, low, medium, high, xhigh, or max. Omit for models with no reasoning knob (Ante then sends no effort parameter and hides the effort slider).
supported_effortsThe effort levels this model accepts on this provider, ascending (same values as effort). Replaces the built-in ladder: the effort slider shows exactly these levels, and OpenAI-compatible providers send the selected level as the same-named reasoning_effort (min is sent as minimal). An empty list declares that the model takes no effort setting. Omit to keep the built-in ladder.
weight_classSize/cost tier — Feather, Middle, or Heavy. Drives automatic default and auxiliary model selection (see Weight classes).
support_visionWhether the model accepts image input. When omitted, Ante infers it from the model id; an image read against a model with no vision support returns the image's metadata instead of its pixels.

The models section patches a model by id wherever it is defined, layering on top of the canonical spec at resolve time. The built-in presets leave it empty:

{
  "models": {
    "claude-sonnet-5": { "effort": "high", "max_tokens": 64000 }
  }
}

Weight classes​

Every built-in model carries a weight_class — Feather (small/fast/cheap), Middle, or Heavy (largest/most capable). Ante uses it to pick models without you naming one:

  • Default model — selecting a provider with no explicit model resolves to its heaviest model.
  • Ambient UI calls — cosmetic thinking phrases and next-prompt suggestions use the provider's lightest model when it is classified as Feather or Middle. WebFetch returns page content directly to the active session model and does not use this path.

Models with no weight_class set fall back to catalog order. Set weight_class on a custom model to steer these choices.

How a model and provider are chosen​

When Ante resolves a model id it layers three sources, last wins:

  1. Canonical spec — the first provider (in catalog order) whose preferred_models lists that id. An unknown id falls back to a bare default spec.
  2. models override — your partial patch for that id, if any.
  3. Environment overrides — the process env vars below win over everything.

Providers are matched to models too. If you pass only --model, Ante picks an authenticated provider whose preferred_models list contains that id, falling back to an authenticated provider with no preferred_models, then to the default provider.

If you pass both --provider and --model, the explicit provider choice wins as long as that provider is authenticated. Ante still resolves the model id globally first, so known ids inherit catalog metadata and user overrides even when served by a different provider; unknown ids are passed through as bare model specs.

When you don't pass --model/--provider, Ante auto-detects: it uses the first authenticated provider in catalog order and that provider's heaviest-weight model (see Weight classes), falling back to local if nothing is authenticated.

Environment overrides​

These process env vars override catalog values for the session without editing the file:

VariableOverrides
MODEL_BASE_URLBase URL for all providers (a per-provider var like ANTHROPIC_BASE_URL takes precedence).
MODEL_TEMPERATUREtemperature
MODEL_TOP_Ptop_p
MODEL_MAX_TOKENSmax_tokens
MODEL_CONTEXT_LIMITcontext_limit

Effort has no env override — set it per run with the --effort CLI flag or per model via the effort field.

Several presets also honor a dedicated base-URL env var that takes precedence over MODEL_BASE_URL: ANTHROPIC_BASE_URL, OPENAI_BASE_URL, OPENAI_COMPATIBLE_BASE_URL, OPENROUTER_BASE_URL, DEEPSEEK_BASE_URL.

Error handling​

A bad entry doesn't sink the whole file. Ante parses catalog.json leniently: any provider or model entry that fails to deserialize is skipped, and the rest still load. Skipped entries are reported at startup, e.g.:

Skipped 1 entry in ~/.ante/catalog.json:
  - providers.my-provider: missing field `base_url`

Only a file that isn't a JSON object at all is rejected outright. A missing catalog.json is fine — Ante just uses the built-in presets.