@tyvm/knowhow 0.0.158 → 0.0.160

package/src/clients/openrouter.ts

@@ -1,24 +0,0 @@
-import { HttpClient, HttpClientOptions } from "./http";
-import { withKnowhowAttribution } from "./attribution";
-import { OpenRouterTextPricing } from "./pricing/openrouter";
-
-/**
- * OpenRouter client — OpenAI-compatible API aggregator
- * https://openrouter.ai/docs
- * 39+ free models; append `:free` suffix to a model id for the free variant.
- * One API key gives access to models from many providers.
- * Set env var OPENROUTER_API_KEY to enable.
- */
-export class GenericOpenRouterClient extends HttpClient {
-  constructor(
-    apiKey = process.env.OPENROUTER_API_KEY,
-    transportOptions: HttpClientOptions = {}
-  ) {
-    super("https://openrouter.ai/api", {
-      ...transportOptions,
-      headers: withKnowhowAttribution(transportOptions.headers),
-    });
-    if (apiKey) this.setJwt(apiKey);
-    this.setPrices(OpenRouterTextPricing);
-  }
-}

package/src/clients/pricing/anthropic.ts

@@ -1,149 +0,0 @@
-/**
- * Anthropic model IDs, pricing, and catalog.
- * Single source of truth for all Anthropic/Claude models.
- * Deprecation dates from: https://docs.anthropic.com/en/docs/about-claude/model-deprecations
- */
-import { completions, ModelCatalogEntry, ModelPricing } from "./types";
-
-// ─── Model IDs ────────────────────────────────────────────────────────────────
-
-export const AnthropicModels = {
-  // Active models
-  Fable5:    "claude-fable-5",
-  // Short aliases (without date suffix) — Anthropic resolves these to the latest
-  // dated version. Pricing entries for the short aliases are added to the pricing
-  // table below so cost tracking works regardless of whether the user specifies
-  // the full dated name or the short alias.
-  Sonnet4_5Short: "claude-sonnet-4-5",
-  Haiku4_5Short:  "claude-haiku-4-5",
-  Opus4_5Short:   "claude-opus-4-5",
-  Sonnet5:   "claude-sonnet-5",
-  Mythos5:   "claude-mythos-5",
-  MythosPreview: "claude-mythos-preview",
-  Opus5:     "claude-opus-5",
-  Opus5Fast: "claude-opus-5-fast",
-  Opus4_8:   "claude-opus-4-8",
-  Opus4_8Fast: "claude-opus-4-8-fast",
-  Opus4_7:   "claude-opus-4-7",
-  Opus4_6:   "claude-opus-4-6",
-  Opus4_6Fast: "claude-opus-4-6-fast",
-  Sonnet4_6: "claude-sonnet-4-6",
-  Opus4_5:   "claude-opus-4-5-20251101",
-  Opus4_1:   "claude-opus-4-1-20250805",
-  Sonnet4_5: "claude-sonnet-4-5-20250929",
-  Haiku4_5:  "claude-haiku-4-5-20251001",
-  // Deprecated models (per Anthropic deprecation page)
-  Opus4:     "claude-opus-4-20250514",   // deprecated 2026-04-14, retirement 2026-06-15
-  Sonnet4:   "claude-sonnet-4-20250514", // deprecated 2026-04-14, retirement 2026-06-15
-  Sonnet3_7: "claude-3-7-sonnet-20250219", // retired 2026-02-19
-  Sonnet3_5: "claude-3-5-sonnet-20241022", // retired 2025-10-28
-  Sonnet3_5_20240620: "claude-3-5-sonnet-20240620", // earlier version, retired
-  Haiku3_5_Latest: "claude-3-5-haiku-latest", // alias → claude-3-5-haiku-20241022, retired
-  Sonnet3:   "claude-3-sonnet-20240229", // retired
-  // models.dev alias IDs (versioned as -0 suffix instead of date)
-  Opus4_0:   "claude-opus-4-0",    // alias for claude-opus-4-20250514, deprecated
-  Sonnet4_0: "claude-sonnet-4-0",  // alias for claude-sonnet-4-20250514, deprecated
-  Haiku3_5:  "claude-3-5-haiku-20241022",  // retired 2026-02-19
-  Opus3:     "claude-3-opus-20240229",   // retired 2026-01-05
-  Haiku3:    "claude-3-haiku-20240307",  // retired 2026-04-20
-} as const;
-
-// ─── Active (non-deprecated) text models ──────────────────────────────────────
-
-export const AnthropicTextModels: string[] = [
-  AnthropicModels.Opus4_8,
-  AnthropicModels.Fable5,
-  AnthropicModels.Sonnet5,
-  AnthropicModels.Opus5,
-  AnthropicModels.Opus4_7,
-  AnthropicModels.Opus4_6,
-  AnthropicModels.Sonnet4_6,
-  AnthropicModels.Opus4_5,
-  AnthropicModels.Opus4_1,
-  AnthropicModels.Sonnet4_5,
-  AnthropicModels.Haiku4_5,
-];
-
-// Models in our catalog but not yet publicly available
-export const AnthropicLimitedAvailabilityModels: string[] = [
-  AnthropicModels.Opus4_8Fast, // fast mode pricing
-  AnthropicModels.Opus5Fast,   // fast mode pricing (research preview)
-  AnthropicModels.Mythos5,     // invitation-only (Project Glasswing)
-  AnthropicModels.MythosPreview, // invitation-only (defensive cybersecurity)
-  AnthropicModels.Opus4_6Fast, // 404 – not publicly available yet
-];
-
-// ─── All models for catalog (active + limited + deprecated/retired) ───────────
-// Deprecation/replacement metadata is embedded in AnthropicTextPricing entries.
-const AnthropicAllModels: string[] = [
-  ...AnthropicTextModels,
-  ...AnthropicLimitedAvailabilityModels,
-  // Short aliases (resolve to their latest dated version) — kept in the catalog
-  // so cost tracking works whether the user passes the alias or the dated id.
-  AnthropicModels.Sonnet4_5Short,
-  AnthropicModels.Haiku4_5Short,
-  AnthropicModels.Opus4_5Short,
-  // Deprecated (retirement 2026-06-15)
-  AnthropicModels.Opus4,
-  AnthropicModels.Sonnet4,
-  // Retired — kept for historical cost tracking
-  AnthropicModels.Sonnet3_7,
-  AnthropicModels.Sonnet3_5,
-  AnthropicModels.Sonnet3_5_20240620,
-  AnthropicModels.Haiku3_5_Latest,
-  AnthropicModels.Sonnet3,
-  AnthropicModels.Opus4_0,
-  AnthropicModels.Sonnet4_0,
-  AnthropicModels.Haiku3_5,
-  AnthropicModels.Opus3,
-  AnthropicModels.Haiku3,
-];
-
-// ─── Pricing (USD per 1M tokens) ──────────────────────────────────────────────
-
-export const AnthropicTextPricing: Record<string, ModelPricing> = {
-  [AnthropicModels.Opus4_8]:   { input: 5.0,  cache_write: 6.25,   cache_hit: 0.5,  output: 25.0 },
-  [AnthropicModels.Opus4_8Fast]: { input: 10.0, cache_write: 12.5,  cache_hit: 1.0,  output: 50.0, limitedAvailability: true },
-  // Claude Opus 5 (launched 2026-06-09) — recommended for complex agentic coding
-  [AnthropicModels.Opus5]:     { input: 5.0,  cache_write: 6.25,   cache_hit: 0.5,  output: 25.0 },
-  [AnthropicModels.Opus5Fast]: { input: 10.0, cache_write: 12.5,   cache_hit: 1.0,  output: 50.0, limitedAvailability: true },
-  // Claude 5 generation (launched 2026-06-09)
-  [AnthropicModels.Fable5]:    { input: 10.0, cache_write: 12.5,   cache_hit: 1.0,  output: 50.0 },
-  [AnthropicModels.Sonnet5]:   { input: 3.0,  cache_write: 3.75,   cache_hit: 0.3,  output: 15.0 },
-  [AnthropicModels.Mythos5]:   { input: 10.0, cache_write: 12.5,   cache_hit: 1.0,  output: 50.0, limitedAvailability: true },
-  [AnthropicModels.MythosPreview]: { input: 10.0, cache_write: 12.5, cache_hit: 1.0, output: 50.0, limitedAvailability: true },
-  [AnthropicModels.Opus4_7]:   { input: 5.0,  cache_write: 6.25,   cache_hit: 0.5,  output: 25.0 },
-  [AnthropicModels.Opus4_6]:   { input: 5.0,  cache_write: 6.25,   cache_hit: 0.5,  output: 25.0 },
-  [AnthropicModels.Opus4_6Fast]: { input: 30.0, cache_write: 37.5, cache_hit: 3.0,  output: 150.0, limitedAvailability: true },
-  [AnthropicModels.Sonnet4_6]: { input: 3.0,  cache_write: 3.75,   cache_hit: 0.3,  output: 15.0 },
-  [AnthropicModels.Opus4_5]:   { input: 5.0,  cache_write: 6.25,   cache_hit: 0.5,  output: 25.0 },
-  [AnthropicModels.Opus4_1]:   { input: 15.0, cache_write: 18.75,  cache_hit: 1.5,  output: 75.0 },
-  [AnthropicModels.Sonnet4_5]: { input: 3.0,  input_gt_200k: 6.0,  cache_write: 3.75, cache_hit: 0.3, output: 15.0, output_gt_200k: 22.5 },
-  [AnthropicModels.Haiku4_5]:  { input: 1.0,  cache_write: 1.25,   cache_hit: 0.1,  output: 5.0 },
-  // Short aliases — same pricing as their dated counterparts
-  [AnthropicModels.Sonnet4_5Short]: { input: 3.0, input_gt_200k: 6.0, cache_write: 3.75, cache_hit: 0.3, output: 15.0, output_gt_200k: 22.5 },
-  [AnthropicModels.Haiku4_5Short]:  { input: 1.0, cache_write: 1.25, cache_hit: 0.1, output: 5.0 },
-  [AnthropicModels.Opus4_5Short]:   { input: 5.0, cache_write: 6.25, cache_hit: 0.5, output: 25.0 },
-  // Deprecated — pricing retained for cost tracking
-  [AnthropicModels.Opus4]:     { input: 15.0, cache_write: 18.75,  cache_hit: 1.5,  output: 75.0,  deprecated: true, deprecationDate: "2026-06-15" },
-  [AnthropicModels.Sonnet4]:   { input: 3.0,  input_gt_200k: 6.0,  cache_write: 3.75, cache_hit: 0.3, output: 15.0, output_gt_200k: 22.5, deprecated: true, deprecationDate: "2026-06-15" },
-  // Retired — pricing retained for cost tracking of historical usage
-  [AnthropicModels.Sonnet3_7]: { input: 3.0,  cache_write: 3.75,   cache_hit: 0.3,  output: 15.0,  deprecated: true, deprecationDate: "2026-02-19" },
-  [AnthropicModels.Sonnet3_5]: { input: 3.0,  cache_write: 3.75,   cache_hit: 0.3,  output: 15.0,  deprecated: true, deprecationDate: "2025-10-28" },
-  [AnthropicModels.Sonnet3_5_20240620]: { input: 3.0, cache_write: 3.75, cache_hit: 0.3, output: 15.0, deprecated: true, deprecationDate: "2025-10-28" },
-  [AnthropicModels.Haiku3_5_Latest]:   { input: 0.8, cache_write: 1.0,   cache_hit: 0.08, output: 4.0,  deprecated: true, deprecationDate: "2026-02-19" },
-  [AnthropicModels.Sonnet3]:           { input: 3.0, cache_write: 3.75,  cache_hit: 0.3,  output: 15.0, deprecated: true },
-  [AnthropicModels.Opus4_0]:           { input: 15.0, cache_write: 18.75, cache_hit: 1.5, output: 75.0, deprecated: true, deprecationDate: "2026-06-15" },
-  [AnthropicModels.Sonnet4_0]:         { input: 3.0,  cache_write: 3.75,  cache_hit: 0.3,  output: 15.0, deprecated: true, deprecationDate: "2026-06-15" },
-  [AnthropicModels.Haiku3_5]:  { input: 0.8,  cache_write: 1.0,    cache_hit: 0.08, output: 4.0,   deprecated: true, deprecationDate: "2026-02-19" },
-  [AnthropicModels.Opus3]:     { input: 15.0, cache_write: 18.75,  cache_hit: 1.5,  output: 75.0,  deprecated: true, deprecationDate: "2026-01-05" },
-  [AnthropicModels.Haiku3]:    { input: 0.25, cache_write: 0.3125, cache_hit: 0.025, output: 1.25, deprecated: true, deprecationDate: "2026-04-20" },
-};
-
-// ─── Catalog ──────────────────────────────────────────────────────────────────
-// Metadata (deprecated, deprecationDate, limitedAvailability) is read directly
-// from the AnthropicTextPricing entries — no need for separate groups.
-
-export const ANTHROPIC_MODEL_CATALOG: ModelCatalogEntry[] = [
-  ...completions(AnthropicAllModels, "anthropic", AnthropicTextPricing),
-];

package/src/clients/pricing/cerebras.ts

@@ -1,11 +0,0 @@
-/**
- * Cerebras pricing (USD per 1M tokens)
- * Source: https://cerebras.ai/pricing
- */
-export const CerebrasTextPricing: Record<string, { input: number; output: number }> = {
-  "llama3.1-8b": { input: 0.10, output: 0.10 },
-  "llama3.3-70b": { input: 0.85, output: 1.20 },
-  "qwen-3-235b-a22b-instruct-2507": { input: 0.60, output: 1.20 },
-  "gpt-oss-120b": { input: 0.25, output: 0.69 },
-  "zai-glm-4.7": { input: 2.25, output: 2.75 },
-};

package/src/clients/pricing/copilot.ts

@@ -1,60 +0,0 @@
-/**
- * GitHub Copilot pricing — "premium request" multiplier system.
- *
- * GitHub Copilot does NOT charge per token. Instead each model call costs a
- * certain number of "premium requests" from the subscriber's monthly allowance.
- * All models are effectively $0/token from the perspective of knowhow pricing,
- * but we store the multiplier as metadata for informational purposes.
- *
- * Copilot Individual: 300 premium requests/month
- * Copilot Pro:        300 premium requests/month
- * Copilot Business:  300 premium requests/month (per seat)
- * Copilot Enterprise: 300 premium requests/month (per seat)
- *
- * Model IDs are as returned by https://api.githubcopilot.com/models
- * Source: https://docs.github.com/en/copilot/reference/ai-models/supported-models
- */
-
-/** Copilot premium request multipliers (informational only) */
-export const CopilotModelMultipliers: Record<string, number> = {
-  // OpenAI models
-  "gpt-4o": 0,          // free (base request)
-  "gpt-4.1": 0,         // free (base request)
-  "gpt-5-mini": 0,      // free (base request)
-  "gpt-5.2": 1,
-  "gpt-5.2-codex": 1,
-  "gpt-5.3-codex": 1,
-  "gpt-5.4": 1,
-  "gpt-5.4-mini": 0.33,
-  "gpt-5.4-nano": 0.25,
-  "gpt-5.5": 7.5,       // promotional until further notice
-  "gpt-5": 1,
-  "gpt-5.1": 1,
-  "gpt-5.1-codex": 1,
-  "gpt-5.1-codex-max": 1,
-  "gpt-5.1-codex-mini": 1,
-  // Anthropic models
-  "claude-haiku-4.5": 0.33,
-  "claude-sonnet-4": 1,
-  "claude-sonnet-4.5": 1,
-  "claude-sonnet-4.6": 1,
-  "claude-opus-4.5": 3,
-  "claude-opus-4.6": 3,
-  "claude-opus-41": 3,
-  "claude-opus-4.7": 7.5, // promotional until April 30, 2026
-  // Google models
-  "gemini-2.5-pro": 1,
-  "gemini-3-flash-preview": 0.33,
-  "gemini-3.1-pro-preview": 1,
-  "gemini-3-pro-preview": 1,
-  // xAI models
-  "grok-code-fast-1": 0.25,
-};
-
-/**
- * Copilot pricing is all $0/token — consumption is via premium request allowance.
- * This map is used by the check-model-pricing script to recognise registered models.
- */
-export const CopilotTextPricing: Record<string, { input: number; output: number }> = Object.fromEntries(
-  Object.keys(CopilotModelMultipliers).map((id) => [id, { input: 0.0, output: 0.0 }])
-);

package/src/clients/pricing/deepseek.ts

@@ -1,15 +0,0 @@
-/**
- * DeepSeek pricing (USD per 1M tokens)
- * Source: https://platform.deepseek.com/pricing
- */
-export const DeepSeekTextPricing: Record<string, { input: number; output: number; cached_input?: number }> = {
-  // DeepSeek V3 / deepseek-chat — updated 2026-04 (now routes to DeepSeek V3-0324)
-  // Source: api-docs.deepseek.com/quick_start/pricing
-  "deepseek-chat": { input: 0.28, output: 0.42, cached_input: 0.028 },
-  // DeepSeek R1 / deepseek-reasoner — same unified pricing as V3
-  // Source: api-docs.deepseek.com/quick_start/pricing (costgoat.com verification)
-  "deepseek-reasoner": { input: 0.28, output: 0.42, cached_input: 0.028 },
-  // DeepSeek V4 Flash & Pro
-  "deepseek-v4-flash": { input: 0.14, output: 0.28 },
-  "deepseek-v4-pro": { input: 1.74, output: 3.48 },
-};

package/src/clients/pricing/fireworks.ts

@@ -1,182 +0,0 @@
-import { ModelPricing, ModelCatalogEntry, completions, embeddings } from "./types";
-
-/**
- * Fireworks AI model IDs, pricing, and catalog.
- * Single source of truth for all Fireworks models.
- * Source: https://fireworks.ai/pricing
- */
-
-// ─── Model IDs ────────────────────────────────────────────────────────────────
-
-export const FireworksModels = {
-  // Moonshot AI
-  KimiK3:      "accounts/fireworks/models/kimi-k3",
-
-  // Meta
-  MuseGlimmer30b: "accounts/fireworks/models/muse-glimmer-30b",
-
-  KimiK2p7Code: "accounts/fireworks/models/kimi-k2p7-code",
-  KimiK2_6:    "accounts/fireworks/models/kimi-k2-6",
-  KimiK2_5:    "accounts/fireworks/models/kimi-k2-5",
-
-  // MiniMax
-  MinimaxM3:   "accounts/fireworks/models/minimax-m3",
-  MinimaxM2_7: "accounts/fireworks/models/minimax-m2-7",
-  MinimaxM2_5: "accounts/fireworks/models/minimax-m2-5",
-
-  // Qwen
-  Qwen3p7Plus: "accounts/fireworks/models/qwen3p7-plus",
-  Qwen3p8Max:  "accounts/fireworks/models/qwen3p8-2p4t-a95b",
-
-  // NVIDIA
-  NemotronLightning3p530bA3b: "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b",
-
-  // Thinking Machines Lab
-  Inkling: "accounts/fireworks/models/inkling",
-
-  // Deprecated Qwen Models
-  Qwen3_6Plus:                "accounts/fireworks/models/qwen3-6-plus",
-  Qwen3p6Plus:                "accounts/fireworks/models/qwen3p6-plus",
-  Qwen3Vl30bA3bThinking:      "accounts/fireworks/models/qwen3-vl-30b-a3b-thinking",
-  Qwen3Vl30bA3bInstruct:      "accounts/fireworks/models/qwen3-vl-30b-a3b-instruct",
-  Qwen3_8b:                   "accounts/fireworks/models/qwen3-8b",
-
-  // Z.ai
-  Glm5p2:  "accounts/fireworks/models/glm-5p2",
-  Glm5p1:  "accounts/fireworks/models/glm-5p1",
-  Glm5:    "accounts/fireworks/models/glm-5",
-  Glm4_7:  "accounts/fireworks/models/glm-4-7",
-
-  // DeepSeek AI
-  DeepseekV4Pro:   "accounts/fireworks/models/deepseek-v4-pro",
-  DeepseekV4Flash: "accounts/fireworks/models/deepseek-v4-flash",
-  DeepseekV3_2:    "accounts/fireworks/models/deepseek-v3-2",
-  DeepseekV3_1:    "accounts/fireworks/models/deepseek-v3-1",
-
-  // OpenAI OSS
-  GptOss120b: "accounts/fireworks/models/gpt-oss-120b",
-
-  // Embedding / Reranker
-  Qwen3Embedding8b: "accounts/fireworks/models/qwen3-embedding-8b",
-  Qwen3Reranker8b:  "accounts/fireworks/models/qwen3-reranker-8b",
-} as const;
-
-// ─── Text model pricing (USD per 1M tokens) ───────────────────────────────────
-
-export const FireworksTextPricing: Record<string, ModelPricing> = {
-  // Moonshot AI
-  [FireworksModels.KimiK3]:       { input: 3.0, cache_hit: 0.30, output: 15.0 },
-  [FireworksModels.KimiK2p7Code]: { input: 0.95, cache_hit: 0.19, output: 4.0 },
-
-  // Meta
-  [FireworksModels.MuseGlimmer30b]: { input: 0.35, cache_hit: 0.04, output: 1.50 },
-
-  [FireworksModels.KimiK2_6]:     { input: 0.95, cache_hit: 0.16, output: 4.0 },
-  [FireworksModels.KimiK2_5]:     { input: 0.60, cache_hit: 0.10, output: 3.0 },
-
-  // MiniMax
-  [FireworksModels.MinimaxM3]:    { input: 0.30, cache_hit: 0.06, output: 1.20 },
-  [FireworksModels.MinimaxM2_7]:  { input: 0.30, cache_hit: 0.06, output: 1.20 },
-  [FireworksModels.MinimaxM2_5]:  { input: 0.30, cache_hit: 0.03, output: 1.20 },
-
-  // Qwen
-  [FireworksModels.Qwen3p7Plus]:  { input: 0.40, cache_hit: 0.08, output: 1.60 },
-
-  [FireworksModels.Qwen3p8Max]:   { input: 2.00, cache_hit: 0.25, output: 6.00 },
-
-  // NVIDIA
-  [FireworksModels.NemotronLightning3p530bA3b]: { input: 0.05, cache_hit: 0.01, output: 0.20 },
-
-  // Thinking Machines Lab
-  [FireworksModels.Inkling]: { input: 1.00, cache_hit: 0.17, output: 4.05 },
-
-  // Deprecated QWEN Models
-  [FireworksModels.Qwen3_6Plus]:           { input: 0.50, cache_hit: 0.10, output: 3.0, deprecated: true },
-  [FireworksModels.Qwen3p6Plus]:           { input: 0.50, cache_hit: 0.10, output: 3.0, deprecated: true },
-  [FireworksModels.Qwen3Vl30bA3bThinking]: { input: 0.15, cache_hit: 0.08, output: 0.60, deprecated: true },
-  [FireworksModels.Qwen3Vl30bA3bInstruct]: { input: 0.15, cache_hit: 0.08, output: 0.60, deprecated: true },
-  [FireworksModels.Qwen3_8b]:              { input: 0.20, cache_hit: 0.10, output: 0.20, deprecated: true },
-
-  // Z.ai
-  [FireworksModels.Glm5p2]: { input: 1.40, cache_hit: 0.26, output: 4.40 },
-  [FireworksModels.Glm5p1]: { input: 1.40, cache_hit: 0.26, output: 4.40 },
-  [FireworksModels.Glm5]:   { input: 1.00, cache_hit: 0.20, output: 3.20, deprecated: true },
-  [FireworksModels.Glm4_7]: { input: 0.60, cache_hit: 0.30, output: 2.20, deprecated: true },
-
-  // DeepSeek AI
-  [FireworksModels.DeepseekV4Pro]:   { input: 1.32, cache_hit: 0.044, output: 3.96 },
-  [FireworksModels.DeepseekV4Flash]: { input: 0.22, cache_hit: 0.007, output: 0.66 },
-  [FireworksModels.DeepseekV3_2]:    { input: 0.56, cache_hit: 0.28, output: 1.68, deprecated: true },
-  [FireworksModels.DeepseekV3_1]:    { input: 0.56, cache_hit: 0.28, output: 1.68, deprecated: true },
-
-  // OpenAI OSS
-  [FireworksModels.GptOss120b]: { input: 0.15, cache_hit: 0.01, output: 0.60 },
-};
-
-/**
- * Fireworks AI embedding/reranker model pricing (USD per 1M tokens)
- * Source: https://fireworks.ai/pricing
- */
-export const FireworksEmbeddingPricing: Record<string, ModelPricing> = {
-  // Qwen3 Embedding 8B — $0.10/M tokens, context 40k
-  [FireworksModels.Qwen3Embedding8b]: { input: 0.10, output: 0.10 },
-
-  // Qwen3 Reranker 8B — $0.20/M tokens, context 40k
-  [FireworksModels.Qwen3Reranker8b]: { input: 0.20, output: 0.20 },
-};
-
-
-// ─── Active text models (non-deprecated) ─────────────────────────────────────
-
-export const FireworksActiveTextModels: string[] = [
-  // Moonshot AI
-  FireworksModels.KimiK3,
-  FireworksModels.KimiK2p7Code,
-  FireworksModels.KimiK2_6,
-  FireworksModels.KimiK2_5,
-  // Meta
-  FireworksModels.MuseGlimmer30b,
-  // MiniMax
-  FireworksModels.MinimaxM3,
-  FireworksModels.MinimaxM2_7,
-  FireworksModels.MinimaxM2_5,
-  // Qwen
-  FireworksModels.Qwen3p7Plus,
-  FireworksModels.Qwen3p8Max,
-  // NVIDIA
-  FireworksModels.NemotronLightning3p530bA3b,
-  // Thinking Machines Lab
-  FireworksModels.Inkling,
-  // Z.ai
-  FireworksModels.Glm5p2,
-  FireworksModels.Glm5p1,
-  // DeepSeek AI
-  FireworksModels.DeepseekV4Pro,
-  FireworksModels.DeepseekV4Flash,
-  // OpenAI OSS
-  FireworksModels.GptOss120b,
-];
-
-export const FireworksDeprecatedTextModels: string[] = [
-  FireworksModels.Qwen3_6Plus,
-  FireworksModels.Qwen3p6Plus,
-  FireworksModels.Qwen3Vl30bA3bThinking,
-  FireworksModels.Qwen3Vl30bA3bInstruct,
-  FireworksModels.Qwen3_8b,
-  FireworksModels.Glm5,
-  FireworksModels.Glm4_7,
-  FireworksModels.DeepseekV3_2,
-  FireworksModels.DeepseekV3_1,
-];
-
-// ─── Catalog ──────────────────────────────────────────────────────────────────
-
-export const FIREWORKS_MODEL_CATALOG: ModelCatalogEntry[] = [
-  ...completions(FireworksActiveTextModels, "fireworks", FireworksTextPricing),
-  ...completions(FireworksDeprecatedTextModels, "fireworks", FireworksTextPricing),
-  ...embeddings(
-    [FireworksModels.Qwen3Embedding8b, FireworksModels.Qwen3Reranker8b],
-    "fireworks",
-    FireworksEmbeddingPricing,
-  ),
-];

package/src/clients/pricing/github.ts

@@ -1,69 +0,0 @@
-/**
- * GitHub Models pricing (USD per 1M tokens)
- * Source: https://github.com/marketplace/models
- * All models are free within rate limits for GitHub users.
- * Low-tier (higher rate limits) and high-tier (lower rate limits) exist.
- */
-export const GitHubModelsTextPricing: Record<string, { input: number; output: number }> = {
-  // OpenAI via GitHub (free)
-  "openai/gpt-4o": { input: 0.0, output: 0.0 },
-  "openai/gpt-4o-mini": { input: 0.0, output: 0.0 },
-  "openai/o1": { input: 0.0, output: 0.0 },
-  "openai/o1-mini": { input: 0.0, output: 0.0 },
-  "openai/o3-mini": { input: 0.0, output: 0.0 },
-  "openai/o4-mini": { input: 0.0, output: 0.0 },
-  "openai/gpt-4.1": { input: 0.0, output: 0.0 },
-  "openai/gpt-4.1-mini": { input: 0.0, output: 0.0 },
-  "openai/gpt-4.1-nano": { input: 0.0, output: 0.0 },
-  "openai/o3": { input: 0.0, output: 0.0 },
-  // DeepSeek via GitHub (free)
-  "deepseek/deepseek-r1": { input: 0.0, output: 0.0 },
-  "deepseek/deepseek-v3-0324": { input: 0.0, output: 0.0 },
-  // Microsoft Phi via GitHub (free)
-  "microsoft/phi-4": { input: 0.0, output: 0.0 },
-  "microsoft/phi-4-mini-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-4-multimodal-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-4-mini-reasoning": { input: 0.0, output: 0.0 },
-  "microsoft/phi-4-reasoning": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3-medium-128k-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3-medium-4k-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3-mini-128k-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3-mini-4k-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3-small-128k-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3-small-8k-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3.5-mini-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3.5-moe-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/phi-3.5-vision-instruct": { input: 0.0, output: 0.0 },
-  "microsoft/mai-ds-r1": { input: 0.0, output: 0.0 },
-  // Meta Llama via GitHub (free)
-  "meta/llama-3.3-70b-instruct": { input: 0.0, output: 0.0 },
-  "meta/llama-3.2-11b-vision-instruct": { input: 0.0, output: 0.0 },
-  "meta/llama-3.1-405b-instruct": { input: 0.0, output: 0.0 },
-  "meta/llama-3.2-90b-vision-instruct": { input: 0.0, output: 0.0 },
-  "meta/llama-4-maverick-17b-128e-instruct-fp8": { input: 0.0, output: 0.0 },
-  "meta/llama-4-scout-17b-16e-instruct": { input: 0.0, output: 0.0 },
-  "meta/meta-llama-3-70b-instruct": { input: 0.0, output: 0.0 },
-  "meta/meta-llama-3-8b-instruct": { input: 0.0, output: 0.0 },
-  "meta/meta-llama-3.1-405b-instruct": { input: 0.0, output: 0.0 },
-  "meta/meta-llama-3.1-70b-instruct": { input: 0.0, output: 0.0 },
-  "meta/meta-llama-3.1-8b-instruct": { input: 0.0, output: 0.0 },
-  // Mistral via GitHub (free)
-  "mistral-ai/mistral-large-2411": { input: 0.0, output: 0.0 },
-  "mistral-ai/mistral-small-2503": { input: 0.0, output: 0.0 },
-  "mistral-ai/codestral-2501": { input: 0.0, output: 0.0 },
-  "mistral-ai/ministral-3b": { input: 0.0, output: 0.0 },
-  "mistral-ai/mistral-medium-2505": { input: 0.0, output: 0.0 },
-  "mistral-ai/mistral-nemo": { input: 0.0, output: 0.0 },
-  // Cohere via GitHub (free)
-  "cohere/cohere-command-r-plus-08-2024": { input: 0.0, output: 0.0 },
-  "cohere/cohere-command-r-08-2024": { input: 0.0, output: 0.0 },
-  "cohere/cohere-command-a": { input: 0.0, output: 0.0 },
-  // AI21 via GitHub (free)
-  "ai21-labs/ai21-jamba-1.5-mini": { input: 0.0, output: 0.0 },
-  "ai21-labs/ai21-jamba-1.5-large": { input: 0.0, output: 0.0 },
-  // Core42
-  "core42/jais-30b-chat": { input: 0.0, output: 0.0 },
-  // xAI via GitHub (free)
-  "xai/grok-3": { input: 0.0, output: 0.0 },
-  "xai/grok-3-mini": { input: 0.0, output: 0.0 },
-};