Module models 

Model catalog — aliases with direct model pinning and optional discovery filters, dependency-tree config merge, and models cache lifecycle.

Model aliases map short names (opus, sonnet, codex) to concrete model IDs. Two modes:

• Pinned: explicit model ID, with optional match/exclude discovery filters.
• AutoResolve: pattern-based resolution against a cached model catalog.

Merge precedence: consumer > deps (declaration order).

Re-exports

pub use availability::ModelAvailability;

Modules

availability

harness

probes

Structs

CachedModel

A single model entry in the cache.

ModelAlias

A model alias — either pinned to a specific model ID or auto-resolved against the models cache at resolution time.

ModelsCache

Cached model catalog from external API.

ResolvedAlias

Fully resolved model alias — everything a consumer needs to launch.

ResolvedDepModels

Info about a resolved dependency’s model config.

Enums

HarnessSource

How the harness was determined.

ModelSpec

How a model alias resolves to a concrete model ID.

RefreshMode

RefreshOutcome

Functions

auto_resolve

Resolve an auto-resolve spec against the models cache.

auto_resolve_all

Resolve an auto-resolve spec against the models cache.

builtin_aliases

Minimal builtin aliases so common model names work out of the box. No descriptions — packages layer those on top. Precedence: consumer > deps > builtins.

ensure_fresh

fetch_models

Fetch models from the models.dev API.

filter_by_visibility

Filter resolved aliases by visibility config.

glob_match

Simple glob matching: * matches any sequence of characters. Everything else is literal. Case-sensitive.

infer_provider_from_model_id

Best-effort provider inference from model ID prefixes. Returns None for unrecognized patterns.

is_mars_offline

load_models_cache_ttl

matches_visibility_pattern

Match a visibility pattern against a resolved model identity.

merge_model_config

Merge model aliases from dependency tree.

now_unix_secs

now_unix_secs_value

read_cache

Read models cache from .mars/models-cache.json.

resolve_all

Resolve all aliases to concrete model IDs + harnesses.

resolve_one

Resolve a single alias and emit diagnostics only for that alias.

resolve_refresh_mode

resolve_with_alias_prefix

Resolve an input like opus-4-6 by matching it against alias filter candidates.

write_cache

Write models cache to .mars/models-cache.json (atomic via tmp+rename).