aigetwey 1.0.1

package/src/providers/capabilities.ts

@@ -0,0 +1,297 @@
+// Model capabilities — what each model can read/do beyond plain text.
+//
+// Fallback order (first match wins), result merged over DEFAULT_CAPABILITIES:
+//   1. PROVIDER_CAPABILITIES[provider][model]  — provider-specific override
+//   2. MODEL_CAPABILITIES[model]               — canonical exact id (handles exceptions)
+//   3. PATTERN_CAPABILITIES                     — glob match, ordered specific -> generic
+//   4. DEFAULT_CAPABILITIES                     — safe floor (always returned)
+//
+// ── HOW TO ADD / UPDATE A MODEL ──────────────────────────────────────
+// Authoritative data source: https://models.dev/api.json (145 providers, 4000+
+// models, MIT). Each model exposes the exact fields we map below:
+//   modalities.input  ["text","image","pdf","audio","video"] -> vision / pdf / audioInput / videoInput
+//   modalities.output ["text","image","audio"]               -> imageOutput / audioOutput
+//   reasoning   -> reasoning      tool_call    -> tools
+//   limit.context -> contextWindow   limit.output -> maxOutput
+// Look up the model id, then:
+//   • If a PATTERN below already covers it correctly -> nothing to do.
+//   • If it is an exception (pattern would mis-match) -> add an exact entry to
+//     MODEL_CAPABILITIES (only the fields that differ from DEFAULT).
+//   • If a whole new family -> add an ordered PATTERN (specific before generic).
+// NOTE: models.dev has NO "search" flag (web search is a runtime tool, not a
+// model spec); set `search` from vendor docs (Claude 4.x+, GPT-5.x/4o, Gemini
+// 2.0+, Grok, Perplexity). Verify with: curl -s https://models.dev/api.json
+
+
+/**
+ * Safe floor — every resolved result is merged over this so consumers
+ * never need null-checks. Most modern LLMs meet these limits.
+ */
+export interface Caps {
+  vision: boolean;
+  pdf: boolean;
+  audioInput: boolean;
+  videoInput: boolean;
+  imageOutput: boolean;
+  audioOutput: boolean;
+  search: boolean;
+  tools: boolean;
+  reasoning: boolean;
+  thinkingFormat: string | null;
+  thinkingCanDisable: boolean;
+  thinkingRange: { min: number; max: number } | null;
+  contextWindow: number;
+  maxOutput: number;
+}
+
+/**
+ * Glob (* = wildcard) match, anchored + case-insensitive. aigetwey's
+ * pricing.matchPattern so capabilities resolve identically.
+ */
+export function matchPattern(pattern: string, model: string): boolean {
+  const regex = new RegExp(
+    "^" + pattern.split("*").map((s) => s.replace(/[.*+?^${}()|[\]\\]/g, "\\$&")).join(".*") + "$",
+    "i",
+  );
+  return regex.test(model);
+}
+
+export const DEFAULT_CAPABILITIES: Caps = {
+  // input modalities
+  vision: false,        // read images
+  pdf: false,           // read PDF / documents
+  audioInput: false,    // read audio
+  videoInput: false,    // read video
+  // output modalities
+  imageOutput: false,   // generate images
+  audioOutput: false,   // generate audio
+  // features
+  search: false,        // built-in web search tool / grounding
+  tools: true,          // function / tool calling
+  reasoning: false,     // thinking / reasoning
+  // thinking wire format (only meaningful when reasoning:true). null → derive from transport.format.
+  // enum: openai|claude-adaptive|claude-budget|gemini-level|gemini-budget|zai|qwen|deepseek|kimi|minimax|hunyuan|step
+  thinkingFormat: null,
+  thinkingCanDisable: true,  // false → model cannot turn thinking off (clamp to min instead of disable)
+  thinkingRange: null,       // { min, max } for budget formats; null = no clamp
+  // limits (tokens)
+  contextWindow: 200000,
+  maxOutput: 64000,
+};
+
+// User-added model metadata can carry dashboard service kinds instead of the
+// runtime capability names used here. Map those typed model kinds into input /
+// output capabilities so custom vision models are not treated as text-only.
+const SERVICE_KIND_CAPABILITIES: Record<string, Partial<Caps>> = {
+  imageToText: { vision: true },
+  image: { imageOutput: true },
+  stt: { audioInput: true },
+  tts: { audioOutput: true },
+  embedding: { tools: false },
+};
+
+export function capabilitiesFromServiceKind(kind: string): Partial<Caps> | null {
+  return SERVICE_KIND_CAPABILITIES[kind] || null;
+}
+
+/**
+ * Canonical exact-id overrides — used for exceptions that patterns would
+ * otherwise mis-match. Only declare deltas vs DEFAULT.
+ */
+export const MODEL_CAPABILITIES: Record<string, Partial<Caps>> = {
+  // Claude 4.6/4.7 have 1M context + adaptive thinking (override generic claude pattern)
+  "claude-opus-4.6":   { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive", contextWindow: 1000000, maxOutput: 128000 },
+  "claude-opus-4.7":   { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive", contextWindow: 1000000, maxOutput: 128000 },
+  "claude-opus-4-6":   { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive", contextWindow: 1000000, maxOutput: 128000 },
+  "claude-sonnet-4.6": { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive", contextWindow: 1000000, maxOutput: 64000 },
+  "claude-sonnet-4-6": { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive", contextWindow: 1000000, maxOutput: 64000 },
+
+  // Gemini image-gen / OpenAI image / xai image variants
+  "gpt-image-1":       { imageOutput: true, tools: false },
+
+  // GLM vision variant (text GLM has no vision)
+  "glm-4.6v":          { vision: true, reasoning: true, thinkingFormat: "zai", contextWindow: 128000 },
+
+  // Qwen plain coder/text (no vision) — registry "vision-model" / "coder-model" aliases
+  "vision-model":      { vision: true, reasoning: true, thinkingFormat: "qwen", contextWindow: 1000000 },
+  "coder-model":       { reasoning: true, thinkingFormat: "qwen", contextWindow: 1000000 },
+};
+
+/**
+ * Provider-specific capability overrides. Keyed by provider alias/id.
+ */
+export const PROVIDER_CAPABILITIES: Record<string, Record<string, Partial<Caps>>> = {
+  // CodeBuddy.cn — authoritative per-model metadata from the gateway's model
+  // config (contextWindow=maxInputTokens, maxOutput=maxOutputTokens, vision=
+  // supportsImages). Every model reasons via OpenAI-style reasoning_effort
+  // (see registry thinkingFormat). `onlyReasoning` models can't turn thinking
+  // off → thinkingCanDisable:false (clamped to minimal instead of disabled).
+  "codebuddy-cn": {
+    "glm-5.2":            { reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 1000000, maxOutput: 48000 },
+    "glm-5.1":            { reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 200000, maxOutput: 48000 },
+    "glm-5.0":            { reasoning: true, thinkingFormat: "openai", contextWindow: 200000, maxOutput: 48000 },
+    "glm-5.0-turbo":      { reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 200000, maxOutput: 48000 },
+    "glm-5v-turbo":       { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 200000, maxOutput: 38000 },
+    "glm-4.7":            { reasoning: true, thinkingFormat: "openai", contextWindow: 200000, maxOutput: 48000 },
+    "minimax-m3":         { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 512000, maxOutput: 48000 },
+    "minimax-m2.7":       { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 200000, maxOutput: 48000 },
+    "kimi-k2.7":          { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 256000, maxOutput: 32000 },
+    "kimi-k2.6":          { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 256000, maxOutput: 32000 },
+    "kimi-k2.5":          { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 164000, maxOutput: 32000 },
+    "hy3-preview":        { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 192000, maxOutput: 64000 },
+    "deepseek-v4-pro":    { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 1000000, maxOutput: 50000 },
+    "deepseek-v4-flash":  { vision: true, reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 1000000, maxOutput: 50000 },
+    "deepseek-v3-2-volc": { reasoning: true, thinkingFormat: "openai", thinkingCanDisable: false, contextWindow: 96000, maxOutput: 32000 },
+  },
+};
+
+/**
+ * Pattern fallback — glob (* = wildcard), matched case-insensitively and
+ * anchored (^...$) so a pattern must match the full model id. ORDER MATTERS:
+ * vision/specific variants first, text-only/generic families last, to avoid
+ * a broad family pattern swallowing an exception (e.g. glm-4.6v vs glm-5).
+ */
+export const PATTERN_CAPABILITIES: Array<{ pattern: string; caps: Partial<Caps> }> = [
+  // ── Claude (4.6+ = adaptive thinking; older/haiku = budget) ──────
+  { pattern: "*claude*opus-4.6*",   caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive" } },
+  { pattern: "*claude*opus-4.7*",   caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive" } },
+  { pattern: "*claude*opus-4.8*",   caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive" } },
+  { pattern: "*claude*sonnet-4.6*", caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive" } },
+  { pattern: "*claude*sonnet-4.7*", caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-adaptive" } },
+  { pattern: "*claude*haiku*",  caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-budget" } },
+  { pattern: "*claude*opus*",   caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-budget" } },
+  { pattern: "*claude*sonnet*", caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-budget" } },
+  { pattern: "*claude*fable*",  caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-budget", contextWindow: 1000000, maxOutput: 128000 } },
+  { pattern: "*claude*mythos*", caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-budget", contextWindow: 1000000, maxOutput: 128000 } },
+  { pattern: "*claude-3*",      caps: { vision: true } },
+  { pattern: "*claude*",        caps: { vision: true, reasoning: true, search: true, thinkingFormat: "claude-budget" } },
+
+  // ── Gemini (all 2.0+ multimodal + google_search grounding, 1M ctx) ─
+  { pattern: "*gemini*image*",  caps: { vision: true, imageOutput: true, contextWindow: 1048576 } },
+  { pattern: "*gemini-3*pro*",  caps: { vision: true, audioInput: true, videoInput: true, reasoning: true, search: true, thinkingFormat: "gemini-level", thinkingCanDisable: false, contextWindow: 1048576, maxOutput: 65535 } },
+  { pattern: "*gemini-3*",      caps: { vision: true, audioInput: true, videoInput: true, reasoning: true, search: true, thinkingFormat: "gemini-level", thinkingCanDisable: false, contextWindow: 1048576, maxOutput: 65536 } },
+  { pattern: "*gemini-2.5*",    caps: { vision: true, audioInput: true, videoInput: true, reasoning: true, search: true, thinkingFormat: "gemini-budget", thinkingRange: { min: 0, max: 24576 }, contextWindow: 1048576, maxOutput: 65536 } },
+  { pattern: "*gemini-2*",      caps: { vision: true, audioInput: true, videoInput: true, search: true, contextWindow: 1048576, maxOutput: 65536 } },
+  { pattern: "*gemini*",        caps: { vision: true, search: true, contextWindow: 1048576 } },
+  { pattern: "*gemma*",         caps: { vision: true, contextWindow: 128000 } },
+  { pattern: "*nanobanana*",    caps: { vision: true, imageOutput: true } },
+
+  // ── OpenAI GPT-5.x (vision + thinking + web search) ──────────────
+  { pattern: "*gpt-5*image*",   caps: { imageOutput: true } },
+  { pattern: "*gpt-5*codex*",   caps: { reasoning: true, search: true, thinkingFormat: "openai", contextWindow: 400000, maxOutput: 128000 } },
+  { pattern: "*gpt-5*",         caps: { vision: true, reasoning: true, search: true, thinkingFormat: "openai", contextWindow: 400000, maxOutput: 128000 } },
+  { pattern: "*gpt-4o*",        caps: { vision: true, search: true, contextWindow: 128000, maxOutput: 16384 } },
+  { pattern: "*gpt-4.1*",       caps: { vision: true, contextWindow: 1000000, maxOutput: 32768 } },
+  { pattern: "*gpt-4-turbo*",   caps: { vision: true, contextWindow: 128000 } },
+  { pattern: "*gpt-4*",         caps: { contextWindow: 128000 } },
+  { pattern: "*gpt-3.5*",       caps: { contextWindow: 16385, maxOutput: 4096 } },
+  { pattern: "*gpt-oss*",       caps: { reasoning: true, thinkingFormat: "openai", contextWindow: 128000 } },
+
+  // ── OpenAI o-series (reasoning, vision) ──────────────────────────
+  { pattern: "*o1-mini*",       caps: { reasoning: true, thinkingFormat: "openai", contextWindow: 128000 } },
+  { pattern: "*o1*",            caps: { vision: true, reasoning: true, thinkingFormat: "openai", contextWindow: 200000, maxOutput: 100000 } },
+  { pattern: "*o3*",            caps: { vision: true, reasoning: true, thinkingFormat: "openai", contextWindow: 200000, maxOutput: 100000 } },
+  { pattern: "*o4*",            caps: { vision: true, reasoning: true, thinkingFormat: "openai", contextWindow: 200000, maxOutput: 100000 } },
+
+  // ── Grok (vision + Live Search) ──────────────────────────────────
+  { pattern: "*grok*image*",    caps: { imageOutput: true } },
+  { pattern: "*grok-code*",     caps: { reasoning: true, thinkingFormat: "openai", contextWindow: 256000 } },
+  { pattern: "*grok-4*",        caps: { vision: true, reasoning: true, search: true, thinkingFormat: "openai", contextWindow: 256000 } },
+  { pattern: "*grok-3*",        caps: { vision: true, reasoning: true, search: true, thinkingFormat: "openai", contextWindow: 131072 } },
+  { pattern: "*grok*",          caps: { vision: true, reasoning: true, search: true, thinkingFormat: "openai", contextWindow: 256000 } },
+
+  // ── Qwen (enable_thinking + thinking_budget; QwQ = thinking-only) ─
+  { pattern: "*qwen*vl*",       caps: { vision: true, reasoning: true, thinkingFormat: "qwen", contextWindow: 262144 } },
+  { pattern: "*qwen*max*",      caps: { vision: true, reasoning: true, thinkingFormat: "qwen", contextWindow: 1000000, maxOutput: 65536 } },
+  { pattern: "*qwen*plus*",     caps: { vision: true, reasoning: true, thinkingFormat: "qwen", contextWindow: 1000000, maxOutput: 65536 } },
+  { pattern: "*qwen*235b*",     caps: { reasoning: true, thinkingFormat: "qwen", contextWindow: 262144 } },
+  { pattern: "*qwen*coder*",    caps: { reasoning: true, thinkingFormat: "qwen", contextWindow: 1000000 } },
+  { pattern: "*qwq*",           caps: { reasoning: true, thinkingFormat: "qwen", thinkingCanDisable: false, contextWindow: 131072 } },
+  { pattern: "*qwen*",          caps: { reasoning: true, thinkingFormat: "qwen", contextWindow: 262144 } },
+
+  // ── Kimi (enabled→reasoning_effort; K2.7-code cannot disable) ─────
+  { pattern: "*kimi*k2.7*code*", caps: { vision: true, reasoning: true, thinkingFormat: "kimi", thinkingCanDisable: false, contextWindow: 262144, maxOutput: 262144 } },
+  { pattern: "*kimi*k2*",       caps: { vision: true, reasoning: true, thinkingFormat: "kimi", contextWindow: 262144, maxOutput: 262144 } },
+  { pattern: "*kimi*",          caps: { reasoning: true, thinkingFormat: "kimi", contextWindow: 262144 } },
+
+  // ── GLM / Z.ai (thinking.enabled; disable via enable_thinking:false) ─
+  { pattern: "*glm-5*",         caps: { reasoning: true, thinkingFormat: "zai", contextWindow: 200000, maxOutput: 128000 } },
+  { pattern: "*glm-4.7*",       caps: { reasoning: true, thinkingFormat: "zai", contextWindow: 200000, maxOutput: 128000 } },
+  { pattern: "*glm-4*",         caps: { reasoning: true, thinkingFormat: "zai", contextWindow: 200000 } },
+  { pattern: "*glm*",           caps: { reasoning: true, thinkingFormat: "zai", contextWindow: 200000 } },
+
+  // ── DeepSeek (thinking.enabled + reasoning_effort; r1 = thinking-only) ─
+  { pattern: "*deepseek-v4*",   caps: { reasoning: true, thinkingFormat: "deepseek", contextWindow: 1000000, maxOutput: 384000 } },
+  { pattern: "*reasoner*",      caps: { reasoning: true, thinkingFormat: "deepseek", thinkingCanDisable: false, contextWindow: 128000 } },
+  { pattern: "*deepseek-r*",    caps: { reasoning: true, thinkingFormat: "deepseek", thinkingCanDisable: false, contextWindow: 128000 } },
+  { pattern: "*deepseek-chat*", caps: { contextWindow: 128000 } },
+  { pattern: "*deepseek*",      caps: { reasoning: true, thinkingFormat: "deepseek", contextWindow: 128000 } },
+
+  // ── MiniMax (M3 = adaptive; M2.x cannot disable) ─────────────────
+  { pattern: "*minimax*image*", caps: { imageOutput: true } },
+  { pattern: "*minimax-m3*",    caps: { vision: true, reasoning: true, thinkingFormat: "minimax", contextWindow: 1048576, maxOutput: 512000 } },
+  { pattern: "*minimax-m2.7*",  caps: { reasoning: true, thinkingFormat: "minimax", thinkingCanDisable: false, contextWindow: 204800, maxOutput: 131072 } },
+  { pattern: "*minimax*",       caps: { reasoning: true, thinkingFormat: "minimax", thinkingCanDisable: false, contextWindow: 200000, maxOutput: 131072 } },
+
+  // ── Xiaomi MiMo (vision, 1M / 262K ctx) ──────────────────────────
+  { pattern: "*mimo*v2.5*",     caps: { vision: true, contextWindow: 1048576, maxOutput: 131072 } },
+  { pattern: "*mimo*omni*",     caps: { vision: true, audioInput: true, contextWindow: 262144, maxOutput: 131072 } },
+  { pattern: "*mimo*",          caps: { vision: true, contextWindow: 262144, maxOutput: 131072 } },
+
+  // ── Llama (4 = vision/1M; 3.x = text-only/128K) ──────────────────
+  { pattern: "*llama-4*",       caps: { vision: true, contextWindow: 1000000 } },
+  { pattern: "*llama*",         caps: { contextWindow: 128000 } },
+
+  // ── Mistral (Large 3 = vision/256K; codestral text) ──────────────
+  { pattern: "*codestral*",     caps: { contextWindow: 256000 } },
+  { pattern: "*mistral-large*", caps: { vision: true, contextWindow: 256000 } },
+  { pattern: "*mistral*",       caps: { contextWindow: 128000 } },
+
+  // ── Cohere (Command A Vision = vision; others text) ──────────────
+  { pattern: "*command-a-vision*", caps: { vision: true, contextWindow: 128000 } },
+  { pattern: "*command*",       caps: { contextWindow: 128000 } },
+
+  // ── Perplexity (web search native) ───────────────────────────────
+  { pattern: "*sonar*",         caps: { search: true, contextWindow: 128000 } },
+  { pattern: "*pplx*",          caps: { search: true, contextWindow: 128000 } },
+  { pattern: "*perplexity*",    caps: { search: true, contextWindow: 128000 } },
+
+  // ── Others ───────────────────────────────────────────────────────
+  { pattern: "*hunyuan*",       caps: { reasoning: true, thinkingFormat: "hunyuan", contextWindow: 262144, maxOutput: 262144 } },
+  { pattern: "hy3*",            caps: { reasoning: true, thinkingFormat: "hunyuan", contextWindow: 262144, maxOutput: 262144 } },
+  { pattern: "*step-*",         caps: { reasoning: true, thinkingFormat: "step", contextWindow: 128000 } },
+  { pattern: "*nemotron*",      caps: { reasoning: true, contextWindow: 128000 } },
+  { pattern: "*ling-*",         caps: { reasoning: true, contextWindow: 128000 } },
+];
+
+/**
+ * Resolve capabilities for a model using the 4-step fallback chain,
+ * merged over DEFAULT_CAPABILITIES so the result is always complete.
+ *
+ * @param {string} provider
+ * @param {string} model
+ * @returns {object} full capabilities object
+ */
+export function getCapabilitiesForModel(provider: string | null, model: string): Caps {
+  if (!model) return { ...DEFAULT_CAPABILITIES };
+
+  // 1. Provider-specific override
+  if (provider && PROVIDER_CAPABILITIES[provider]?.[model]) {
+    return { ...DEFAULT_CAPABILITIES, ...PROVIDER_CAPABILITIES[provider][model] };
+  }
+
+  // 2. Canonical exact (strip vendor prefix: "anthropic/claude-opus-4.7" -> "claude-opus-4.7")
+  const baseModel = (model.includes("/") ? model.split("/").pop() : model) ?? model;
+  if (MODEL_CAPABILITIES[baseModel]) return { ...DEFAULT_CAPABILITIES, ...MODEL_CAPABILITIES[baseModel] };
+  if (MODEL_CAPABILITIES[model]) return { ...DEFAULT_CAPABILITIES, ...MODEL_CAPABILITIES[model] };
+
+  // 3. Pattern match (first match wins)
+  for (const { pattern, caps } of PATTERN_CAPABILITIES) {
+    if (matchPattern(pattern, baseModel) || matchPattern(pattern, model)) {
+      return { ...DEFAULT_CAPABILITIES, ...caps };
+    }
+  }
+
+  // 4. Floor
+  return { ...DEFAULT_CAPABILITIES };
+}

package/src/providers/free.ts

@@ -0,0 +1,53 @@
+/**
+ * Free passthrough providers (e.g. OpenCode Free). These speak OpenAI format,
+ * need no upstream auth, and expose their catalog at `{base_url}/models`. The
+ * gateway already routes them through the normal pipeline (the key pool hands
+ * out an empty key, the client omits the auth header); this module only adds the
+ * one extra capability they need — fetching the model list at runtime so it
+ * doesn't have to be hand-maintained in config.
+ */
+import { request } from "undici";
+import type { Provider } from "../config.js";
+
+export interface FetchedModel {
+  id: string;
+}
+
+export interface ModelFetchResult {
+  ok: boolean;
+  models: FetchedModel[];
+  error?: string;
+}
+
+/**
+ * Fetch a provider's model catalog from `{base_url}/models` (OpenAI shape:
+ * `{ data: [{ id }] }`). Never throws — returns a structured result so the
+ * dashboard's "Connect OpenCode Free" button can surface failures inline.
+ */
+export async function fetchModels(provider: Provider): Promise<ModelFetchResult> {
+  const base = provider.base_url.replace(/\/$/, "");
+  const headers: Record<string, string> = { ...(provider.headers ?? {}) };
+  // a free provider has no key; a keyed one still gets its bearer for /models.
+  const key = provider.api_keys?.[0] ?? provider.api_key;
+  if (key && !provider.free) headers["authorization"] = `Bearer ${key}`;
+
+  try {
+    const res = await request(`${base}/models`, {
+      method: "GET",
+      headers,
+      headersTimeout: 15_000,
+      bodyTimeout: 15_000,
+    });
+    if (res.statusCode >= 400) {
+      await res.body.dump();
+      return { ok: false, models: [], error: `models endpoint returned ${res.statusCode}` };
+    }
+    const body = (await res.body.json()) as { data?: Array<{ id?: unknown }> };
+    const models = (body.data ?? [])
+      .map((m) => (typeof m.id === "string" ? { id: m.id } : null))
+      .filter((m): m is FetchedModel => m !== null);
+    return { ok: true, models };
+  } catch (e) {
+    return { ok: false, models: [], error: (e as Error).message };
+  }
+}

package/src/providers/pricing.ts

@@ -0,0 +1,261 @@
+// Pricing rates for AI models — all rates in $/1M tokens
+//
+// Fallback order (first match wins):
+//   1. PROVIDER_PRICING[provider][model]  — provider-specific override
+//   2. MODEL_PRICING[model]               — canonical model price (provider-agnostic)
+//   3. PATTERN_PRICING                    — glob pattern match (e.g. "codex-*")
+
+/**
+ * All rates in $/1M tokens. Implemented for aigetwey so cost auto-resolves per model the same way. See
+ * 
+ */
+export interface Pricing {
+  input: number;
+  output: number;
+  cached?: number;
+  reasoning?: number;
+  cache_creation?: number;
+}
+
+/**
+ * Canonical model pricing — provider-agnostic.
+ * Cover all known models; deduplicated across providers.
+ */
+export const MODEL_PRICING: Record<string, Pricing> = {
+  // === Anthropic / Claude ===
+  "claude-opus-4-6":              { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 25.00,  cache_creation: 6.25  },
+  "claude-opus-4-5-20251101":     { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 25.00,  cache_creation: 6.25  },
+  "claude-sonnet-4-6":            { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 15.00,  cache_creation: 3.75  },
+  "claude-sonnet-4-5-20250929":   { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 15.00,  cache_creation: 3.75  },
+  "claude-haiku-4-5-20251001":    { input: 1.00,  output: 5.00,  cached: 0.10,  reasoning: 5.00,   cache_creation: 1.25  },
+  "claude-sonnet-4-20250514":     { input: 3.00,  output: 15.00, cached: 1.50,  reasoning: 15.00,  cache_creation: 3.00  },
+  "claude-opus-4-20250514":       { input: 15.00, output: 25.00, cached: 7.50,  reasoning: 112.50, cache_creation: 15.00 },
+  "claude-3-5-sonnet-20241022":   { input: 3.00,  output: 15.00, cached: 1.50,  reasoning: 15.00,  cache_creation: 3.00  },
+  "claude-haiku-4.5":             { input: 0.50,  output: 2.50,  cached: 0.05,  reasoning: 3.75,   cache_creation: 0.50  },
+  "claude-opus-4.1":              { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+  "claude-opus-4.5":              { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+  "claude-opus-4.6":              { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+  "claude-sonnet-4":              { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 22.50,  cache_creation: 3.00  },
+  "claude-sonnet-4.5":            { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 22.50,  cache_creation: 3.00  },
+  "claude-sonnet-4.6":            { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 22.50,  cache_creation: 3.00  },
+  "claude-opus-4-5-thinking":     { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+  "claude-opus-4-6-thinking":     { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 37.50,  cache_creation: 5.00  },
+
+  // === OpenAI / GPT ===
+  "gpt-3.5-turbo":                { input: 0.50,  output: 1.50,  cached: 0.25,  reasoning: 2.25,   cache_creation: 0.50  },
+  "gpt-4":                        { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  },
+  "gpt-4-turbo":                  { input: 10.00, output: 30.00, cached: 5.00,  reasoning: 45.00,  cache_creation: 10.00 },
+  "gpt-4o":                       { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  },
+  "gpt-4o-mini":                  { input: 0.15,  output: 0.60,  cached: 0.075, reasoning: 0.90,   cache_creation: 0.15  },
+  "gpt-4.1":                      { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  },
+  "gpt-5":                        { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  },
+  "gpt-5-mini":                   { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  },
+  "gpt-5-codex":                  { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  },
+  "gpt-5.1":                      { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  },
+  "gpt-5.1-codex":                { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  },
+  "gpt-5.1-codex-mini":           { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  },
+  "gpt-5.1-codex-mini-high":      { input: 2.00,  output: 8.00,  cached: 1.00,  reasoning: 12.00,  cache_creation: 2.00  },
+  "gpt-5.1-codex-max":            { input: 8.00,  output: 32.00, cached: 4.00,  reasoning: 48.00,  cache_creation: 8.00  },
+  "gpt-5.2":                      { input: 5.00,  output: 20.00, cached: 2.50,  reasoning: 30.00,  cache_creation: 5.00  },
+  "gpt-5.2-codex":                { input: 5.00,  output: 20.00, cached: 2.50,  reasoning: 30.00,  cache_creation: 5.00  },
+  "gpt-5.3-codex":                { input: 6.00,  output: 24.00, cached: 3.00,  reasoning: 36.00,  cache_creation: 6.00  },
+  "gpt-5.3-codex-xhigh":         { input: 10.00, output: 40.00, cached: 5.00,  reasoning: 60.00,  cache_creation: 10.00 },
+  "gpt-5.3-codex-high":          { input: 8.00,  output: 32.00, cached: 4.00,  reasoning: 48.00,  cache_creation: 8.00  },
+  "gpt-5.3-codex-low":           { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  },
+  "gpt-5.3-codex-none":          { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  },
+  "gpt-5.3-codex-spark":         { input: 3.00,  output: 12.00, cached: 0.30,  reasoning: 12.00,  cache_creation: 3.00  },
+  "o1":                           { input: 15.00, output: 60.00, cached: 7.50,  reasoning: 90.00,  cache_creation: 15.00 },
+  "o1-mini":                      { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  },
+
+  // === Gemini ===
+  "gemini-3-flash-preview":       { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  },
+  "gemini-3-pro-preview":         { input: 2.00,  output: 12.00, cached: 0.25,  reasoning: 18.00,  cache_creation: 2.00  },
+  "gemini-3.1-pro-low":           { input: 2.00,  output: 12.00, cached: 0.25,  reasoning: 18.00,  cache_creation: 2.00  },
+  "gemini-3.1-pro-high":          { input: 4.00,  output: 18.00, cached: 0.50,  reasoning: 27.00,  cache_creation: 4.00  },
+  "gemini-pro-agent":             { input: 4.00,  output: 18.00, cached: 0.50,  reasoning: 27.00,  cache_creation: 4.00  },
+  "gemini-3-flash-agent":         { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  },
+  "gemini-3.5-flash-low":         { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  },
+  "gemini-3.5-flash-extra-low":   { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  },
+  "gemini-3-flash":               { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  },
+  "gemini-2.5-pro":               { input: 2.00,  output: 12.00, cached: 0.25,  reasoning: 18.00,  cache_creation: 2.00  },
+  "gemini-2.5-flash":             { input: 0.30,  output: 2.50,  cached: 0.03,  reasoning: 3.75,   cache_creation: 0.30  },
+  "gemini-2.5-flash-lite":        { input: 0.15,  output: 1.25,  cached: 0.015, reasoning: 1.875,  cache_creation: 0.15  },
+
+  // === Qwen ===
+  "qwen3-coder-plus":             { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+  "qwen3-coder-flash":            { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+
+  // === Kimi ===
+  "kimi-k2":                      { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+  "kimi-k2-thinking":             { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  },
+  "kimi-k2.5":                    { input: 1.20,  output: 4.80,  cached: 0.60,  reasoning: 7.20,   cache_creation: 1.20  },
+  "kimi-k2.5-thinking":           { input: 1.80,  output: 7.20,  cached: 0.90,  reasoning: 10.80,  cache_creation: 1.80  },
+  "kimi-latest":                  { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+
+  // === DeepSeek ===
+  "deepseek-chat":                { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  },
+  "deepseek-reasoner":            { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  },
+  "deepseek-r1":                  { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  },
+  "deepseek-v3.2-chat":           { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  },
+  "deepseek-v3.2-reasoner":       { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  },
+  "deepseek-v4-flash":            { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  },
+  "deepseek-v4-pro":              { input: 0.435, output: 0.87,  cached: 0.003625, reasoning: 0.87,  cache_creation: 0.435 },
+
+  // === GLM ===
+  "glm-4.6":                      { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "glm-4.6v":                     { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  },
+  "glm-4.7":                      { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  },
+  "glm-5":                        { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+
+  // === MiniMax ===
+  "MiniMax-M3":                   { input: 0.30,  output: 1.20,  cached: 0.06,  reasoning: 1.80,   cache_creation: 0.30  },
+  "MiniMax-M2.1":                 { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "MiniMax-M2.5":                 { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "MiniMax-M2.7":                 { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "minimax-m2.1":                 { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "minimax-m2.5":                 { input: 0.60,  output: 2.40,  cached: 0.30,  reasoning: 3.60,   cache_creation: 0.60  },
+
+  // === Grok ===
+  "grok-code-fast-1":             { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+
+  // === OpenRouter fallback ===
+  "auto":                         { input: 2.00,  output: 8.00,  cached: 1.00,  reasoning: 12.00,  cache_creation: 2.00  },
+
+  // === Misc ===
+  "oswe-vscode-prime":            { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  },
+  "gpt-oss-120b-medium":          { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  },
+  "vision-model":                 { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  },
+  "coder-model":                  { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  },
+};
+
+/**
+ * Provider-specific pricing overrides.
+ * Only include entries where price DIFFERS from MODEL_PRICING.
+ * Keyed by provider alias (cc, cx, gc, gh, ...) or provider id (openai, anthropic, ...).
+ */
+export const PROVIDER_PRICING: Record<string, Record<string, Pricing>> = {
+  // GitHub Copilot (gh) — gpt-5.3-codex has different rate than canonical
+  gh: {
+    "gpt-5.3-codex": { input: 1.75, output: 14.00, cached: 0.175, reasoning: 14.00, cache_creation: 1.75 },
+  },
+};
+
+/**
+ * Pattern-based pricing fallback — matched when no exact model entry found.
+ * Patterns use simple glob: "*" matches any substring.
+ * First match wins — order matters.
+ */
+export const PATTERN_PRICING: Array<{ pattern: string; pricing: Pricing }> = [
+  // --- Codex variants ---
+  { pattern: "*-codex-xhigh",   pricing: { input: 10.00, output: 40.00, cached: 5.00,  reasoning: 60.00,  cache_creation: 10.00 } },
+  { pattern: "*-codex-high",    pricing: { input: 8.00,  output: 32.00, cached: 4.00,  reasoning: 48.00,  cache_creation: 8.00  } },
+  { pattern: "*-codex-max",     pricing: { input: 8.00,  output: 32.00, cached: 4.00,  reasoning: 48.00,  cache_creation: 8.00  } },
+  { pattern: "*-codex-mini-*",  pricing: { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  } },
+  { pattern: "*-codex-mini",    pricing: { input: 1.50,  output: 6.00,  cached: 0.75,  reasoning: 9.00,   cache_creation: 1.50  } },
+  { pattern: "*-codex-low",     pricing: { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  } },
+  { pattern: "*-codex-none",    pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "*-codex-spark",   pricing: { input: 3.00,  output: 12.00, cached: 0.30,  reasoning: 12.00,  cache_creation: 3.00  } },
+  { pattern: "codex-*",         pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "*-codex",         pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+
+  // --- Claude ---
+  { pattern: "claude-opus-*",   pricing: { input: 5.00,  output: 25.00, cached: 0.50,  reasoning: 25.00,  cache_creation: 6.25  } },
+  { pattern: "claude-sonnet-*", pricing: { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 15.00,  cache_creation: 3.75  } },
+  { pattern: "claude-haiku-*",  pricing: { input: 1.00,  output: 5.00,  cached: 0.10,  reasoning: 5.00,   cache_creation: 1.25  } },
+  { pattern: "claude-*",        pricing: { input: 3.00,  output: 15.00, cached: 0.30,  reasoning: 15.00,  cache_creation: 3.75  } },
+
+  // --- Gemini (specific first, generic last) ---
+  { pattern: "gemini-*-flash-lite", pricing: { input: 0.15, output: 1.25, cached: 0.015, reasoning: 1.875, cache_creation: 0.15 } },
+  { pattern: "gemini-*-flash",  pricing: { input: 0.30,  output: 2.50,  cached: 0.03,  reasoning: 3.75,   cache_creation: 0.30  } },
+  { pattern: "gemini-*-pro",    pricing: { input: 2.00,  output: 12.00, cached: 0.25,  reasoning: 18.00,  cache_creation: 2.00  } },
+  { pattern: "gemini-3-*",      pricing: { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  } },
+  { pattern: "gemini-2.5-*",    pricing: { input: 0.30,  output: 2.50,  cached: 0.03,  reasoning: 3.75,   cache_creation: 0.30  } },
+  { pattern: "gemini-*",        pricing: { input: 0.50,  output: 3.00,  cached: 0.03,  reasoning: 4.50,   cache_creation: 0.50  } },
+
+  // --- GPT (specific first, generic last) ---
+  { pattern: "gpt-5.3-*",       pricing: { input: 6.00,  output: 24.00, cached: 3.00,  reasoning: 36.00,  cache_creation: 6.00  } },
+  { pattern: "gpt-5.2-*",       pricing: { input: 5.00,  output: 20.00, cached: 2.50,  reasoning: 30.00,  cache_creation: 5.00  } },
+  { pattern: "gpt-5.1-*",       pricing: { input: 4.00,  output: 16.00, cached: 2.00,  reasoning: 24.00,  cache_creation: 4.00  } },
+  { pattern: "gpt-5-*",         pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "gpt-5*",          pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "gpt-4o-*",        pricing: { input: 0.15,  output: 0.60,  cached: 0.075, reasoning: 0.90,   cache_creation: 0.15  } },
+  { pattern: "gpt-4o",          pricing: { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  } },
+  { pattern: "gpt-4*",          pricing: { input: 2.50,  output: 10.00, cached: 1.25,  reasoning: 15.00,  cache_creation: 2.50  } },
+
+  // --- o1 / o-series ---
+  { pattern: "o1-*",            pricing: { input: 3.00,  output: 12.00, cached: 1.50,  reasoning: 18.00,  cache_creation: 3.00  } },
+  { pattern: "o1",              pricing: { input: 15.00, output: 60.00, cached: 7.50,  reasoning: 90.00,  cache_creation: 15.00 } },
+  { pattern: "o3-*",            pricing: { input: 10.00, output: 40.00, cached: 5.00,  reasoning: 60.00,  cache_creation: 10.00 } },
+  { pattern: "o4-*",            pricing: { input: 2.00,  output: 8.00,  cached: 1.00,  reasoning: 12.00,  cache_creation: 2.00  } },
+
+  // --- Qwen ---
+  { pattern: "qwen3-coder-*",   pricing: { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  } },
+  { pattern: "qwen*-coder-*",   pricing: { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  } },
+  { pattern: "qwen*",           pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+
+  // --- Kimi ---
+  { pattern: "kimi-*-thinking",  pricing: { input: 1.80,  output: 7.20,  cached: 0.90,  reasoning: 10.80,  cache_creation: 1.80  } },
+  { pattern: "kimi-k2*",        pricing: { input: 1.20,  output: 4.80,  cached: 0.60,  reasoning: 7.20,   cache_creation: 1.20  } },
+  { pattern: "kimi-*",          pricing: { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  } },
+
+  // --- DeepSeek ---
+  { pattern: "deepseek-*reasoner*", pricing: { input: 0.14, output: 0.28, cached: 0.0028, reasoning: 0.28, cache_creation: 0.14 } },
+  { pattern: "deepseek-r*",     pricing: { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  } },
+  { pattern: "deepseek-v*",     pricing: { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  } },
+  { pattern: "deepseek-*",      pricing: { input: 0.14,  output: 0.28,  cached: 0.0028, reasoning: 0.28,   cache_creation: 0.14  } },
+
+  // --- GLM ---
+  { pattern: "glm-5*",          pricing: { input: 1.00,  output: 4.00,  cached: 0.50,  reasoning: 6.00,   cache_creation: 1.00  } },
+  { pattern: "glm-4*",          pricing: { input: 0.75,  output: 3.00,  cached: 0.375, reasoning: 4.50,   cache_creation: 0.75  } },
+  { pattern: "glm-*",           pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+
+  // --- MiniMax ---
+  { pattern: "MiniMax-*",       pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+  { pattern: "minimax-*",       pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+
+  // --- Grok ---
+  { pattern: "grok-code-*",     pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+  { pattern: "grok-*",          pricing: { input: 0.50,  output: 2.00,  cached: 0.25,  reasoning: 3.00,   cache_creation: 0.50  } },
+];
+
+/**
+ * Match a model ID against a glob pattern (* = wildcard). Case-insensitive:
+ * registry ids mix casing (e.g. "MiniMax-M2.5" vs "minimax-m2.5").
+ */
+export function matchPattern(pattern: string, model: string): boolean {
+  const regex = new RegExp("^" + pattern.split("*").map(s => s.replace(/[.*+?^${}()|[\]\\]/g, "\\$&")).join(".*") + "$", "i");
+  return regex.test(model);
+}
+
+/**
+ * Resolve pricing for a model using the 3-step fallback chain:
+ *   1. PROVIDER_PRICING[provider][model]
+ *   2. MODEL_PRICING[model]
+ *   3. PATTERN_PRICING (glob match)
+ *
+ * @param {string} provider
+ * @param {string} model
+ * @returns {object|null}
+ */
+export function getPricingForModel(provider: string | null, model: string): Pricing | null {
+  if (!model) return null;
+
+  // 1. Provider-specific override
+  if (provider && PROVIDER_PRICING[provider]?.[model]) {
+    return PROVIDER_PRICING[provider][model];
+  }
+
+  // 2. Canonical model pricing (strip vendor prefix if needed: "deepseek/deepseek-chat" → "deepseek-chat")
+  const baseModel = (model.includes("/") ? model.split("/").pop() : model) ?? model;
+  if (MODEL_PRICING[baseModel]) return MODEL_PRICING[baseModel];
+  if (MODEL_PRICING[model]) return MODEL_PRICING[model];
+
+  // 3. Pattern match
+  for (const { pattern, pricing } of PATTERN_PRICING) {
+    if (matchPattern(pattern, baseModel) || matchPattern(pattern, model)) {
+      return pricing;
+    }
+  }
+
+  return null;
+}