@pydantic/genai-prices 0.0.67 → 0.0.69

package/dist/index.js

@@ -621,6 +621,51 @@ const G = [
           }
         ]
       },
+      {
+        id: "claude-sonnet-5",
+        name: "Claude Sonnet 5",
+        description: "Our most agentic Sonnet model, approaching Opus 4.8 capability at lower cost",
+        match: {
+          or: [
+            {
+              starts_with: "claude-sonnet-5"
+            },
+            {
+              starts_with: "claude-sonnet-5.0"
+            },
+            {
+              starts_with: "claude-5-sonnet"
+            },
+            {
+              starts_with: "claude-5.0-sonnet"
+            }
+          ]
+        },
+        context_window: 1e6,
+        price_comments: "Flat pricing across full 1M context window (no tiered pricing). Introductory pricing ($2/$10 per MTok) applies through 2026-08-31; standard pricing ($3/$15) applies from 2026-09-01. Ref: https://www.anthropic.com/news/claude-sonnet-5",
+        prices: [
+          {
+            prices: {
+              input_mtok: 2,
+              cache_write_mtok: 2.5,
+              cache_read_mtok: 0.2,
+              output_mtok: 10
+            }
+          },
+          {
+            constraint: {
+              start_date: "2026-09-01",
+              type: "start_date"
+            },
+            prices: {
+              input_mtok: 3,
+              cache_write_mtok: 3.75,
+              cache_read_mtok: 0.3,
+              output_mtok: 15
+            }
+          }
+        ]
+      },
       {
         id: "claude-v1",
         description: "Retired, here to match price sources",
@@ -1058,6 +1103,35 @@ const G = [
           }
         }
       },
+      {
+        id: "global.anthropic.claude-sonnet-5-v1:0",
+        match: {
+          contains: "global.anthropic.claude-sonnet-5"
+        },
+        price_comments: "Flat pricing across full 1M context window (no tiered pricing). Promotional launch pricing ($2/$10 per MTok) through 2026-08-31; standard ($3/$15) from 2026-09-01. Ref: https://aws.amazon.com/bedrock/pricing/",
+        prices: [
+          {
+            prices: {
+              input_mtok: 2,
+              cache_write_mtok: 2.5,
+              cache_read_mtok: 0.2,
+              output_mtok: 10
+            }
+          },
+          {
+            constraint: {
+              start_date: "2026-09-01",
+              type: "start_date"
+            },
+            prices: {
+              input_mtok: 3,
+              cache_write_mtok: 3.75,
+              cache_read_mtok: 0.3,
+              output_mtok: 15
+            }
+          }
+        ]
+      },
       {
         id: "google.gemma-3-12b-it",
         name: "Gemma 3 12B IT",
@@ -1388,6 +1462,30 @@ const G = [
           output_mtok: 0.65
         }
       },
+      {
+        id: "openai.gpt-5.4",
+        name: "GPT-5.4",
+        match: {
+          equals: "openai.gpt-5.4"
+        },
+        prices: {
+          input_mtok: 2.75,
+          cache_read_mtok: 0.275,
+          output_mtok: 16.5
+        }
+      },
+      {
+        id: "openai.gpt-5.5",
+        name: "GPT-5.5",
+        match: {
+          equals: "openai.gpt-5.5"
+        },
+        prices: {
+          input_mtok: 5.5,
+          cache_read_mtok: 0.55,
+          output_mtok: 33
+        }
+      },
       {
         id: "openai.gpt-oss-120b-1:0",
         name: "gpt-oss-120b",
@@ -2015,6 +2113,60 @@ const G = [
             ]
           }
         }
+      },
+      {
+        id: "regional.anthropic.claude-sonnet-5-v1:0",
+        match: {
+          or: [
+            {
+              starts_with: "anthropic.claude-sonnet-5"
+            },
+            {
+              starts_with: "claude-sonnet-5"
+            },
+            {
+              contains: "us.anthropic.claude-sonnet-5"
+            },
+            {
+              contains: "au.anthropic.claude-sonnet-5"
+            },
+            {
+              contains: "apac.anthropic.claude-sonnet-5"
+            },
+            {
+              contains: "eu.anthropic.claude-sonnet-5"
+            },
+            {
+              contains: "us-gov.anthropic.claude-sonnet-5"
+            },
+            {
+              contains: "jp.anthropic.claude-sonnet-5"
+            }
+          ]
+        },
+        price_comments: "Regional/cross-region endpoints carry a 10% premium over global (AWS published only the global promo rate; regional computed as global +10%, per the documented regional premium). Promotional launch pricing through 2026-08-31; standard from 2026-09-01. Ref: https://aws.amazon.com/bedrock/pricing/",
+        prices: [
+          {
+            prices: {
+              input_mtok: 2.2,
+              cache_write_mtok: 2.75,
+              cache_read_mtok: 0.22,
+              output_mtok: 11
+            }
+          },
+          {
+            constraint: {
+              start_date: "2026-09-01",
+              type: "start_date"
+            },
+            prices: {
+              input_mtok: 3.3,
+              cache_write_mtok: 4.125,
+              cache_read_mtok: 0.33,
+              output_mtok: 16.5
+            }
+          }
+        ]
       }
     ]
   },
@@ -3307,6 +3459,20 @@ Compared to other leading proprietary and open-weights models Command A delivers
           output_mtok: 4.4
         }
       },
+      {
+        id: "glm-5p2",
+        name: "GLM-5.2",
+        description: "GLM-5.2 introduces a robust 1M-token context and advanced, multi-effort coding capabilities to significantly enhance performance on long-horizon tasks. Features a new IndexShare architecture and improved MTP layer for greater efficiency. 743B parameter MoE model from Z.ai.",
+        match: {
+          equals: "accounts/fireworks/models/glm-5p2"
+        },
+        context_window: 104e4,
+        prices: {
+          input_mtok: 1.4,
+          cache_read_mtok: 0.14,
+          output_mtok: 4.4
+        }
+      },
       {
         id: "gpt-oss-120b",
         name: "OpenAI gpt-oss-120b",
@@ -3361,6 +3527,20 @@ Compared to other leading proprietary and open-weights models Command A delivers
           output_mtok: 4
         }
       },
+      {
+        id: "kimi-k2p7-code",
+        name: "Kimi K2.7 Code",
+        description: "Kimi K2.7 Code is a coding-focused agentic model built upon Kimi K2.6, delivering substantial improvements on real-world long-horizon coding tasks while reducing thinking tokens by roughly 30% compared to its predecessor.",
+        match: {
+          equals: "accounts/fireworks/models/kimi-k2p7-code"
+        },
+        context_window: 262144,
+        prices: {
+          input_mtok: 0.95,
+          cache_read_mtok: 0.19,
+          output_mtok: 4
+        }
+      },
       {
         id: "llama-v3p1-8b-instruct",
         name: "Llama 3.1 8B Instruct",
@@ -3427,6 +3607,20 @@ Compared to other leading proprietary and open-weights models Command A delivers
           output_mtok: 1.2
         }
       },
+      {
+        id: "nemotron-3-ultra-nvfp4",
+        name: "NVIDIA Nemotron 3 Ultra NVFP4",
+        description: "Frontier-scale LLM from NVIDIA using a hybrid Latent Mixture-of-Experts (LatentMoE) architecture with interleaved Mamba-2 and MoE layers plus select Attention layers. Features 55B active parameters out of 550B total and Multi-Token Prediction layers for faster generation, optimized for complex multi-step agents, long-context analysis, and high-accuracy reasoning over code, math, and science.",
+        match: {
+          equals: "accounts/fireworks/models/nemotron-3-ultra-nvfp4"
+        },
+        context_window: 262e3,
+        prices: {
+          input_mtok: 0.6,
+          cache_read_mtok: 0.12,
+          output_mtok: 2.4
+        }
+      },
       {
         id: "qwen2p5-vl-72b-instruct",
         name: "Qwen2.5-VL 72B Instruct",
@@ -3465,6 +3659,20 @@ Compared to other leading proprietary and open-weights models Command A delivers
           cache_read_mtok: 0.1,
           output_mtok: 3
         }
+      },
+      {
+        id: "qwen3p7-plus",
+        name: "Qwen3.7 Plus",
+        description: "Qwen3.7 Plus is Alibaba's latest flagship closed model, available exclusively through Fireworks AI outside of Alibaba's own infrastructure.",
+        match: {
+          equals: "accounts/fireworks/models/qwen3p7-plus"
+        },
+        context_window: 262144,
+        prices: {
+          input_mtok: 0.4,
+          cache_read_mtok: 0.08,
+          output_mtok: 1.6
+        }
       }
     ]
   },
@@ -8876,6 +9084,65 @@ Compared to other leading proprietary and open-weights models Command A delivers
       ]
     },
     extractors: [
+      {
+        api_flavor: "default",
+        root: "usage",
+        model_path: "model",
+        mappings: [
+          {
+            path: "input_tokens",
+            dest: "input_tokens",
+            required: !0
+          },
+          {
+            path: "cache_creation_input_tokens",
+            dest: "input_tokens",
+            required: !1
+          },
+          {
+            path: "cache_read_input_tokens",
+            dest: "input_tokens",
+            required: !1
+          },
+          {
+            path: "cache_creation_input_tokens",
+            dest: "cache_write_tokens",
+            required: !1
+          },
+          {
+            path: "cache_read_input_tokens",
+            dest: "cache_read_tokens",
+            required: !1
+          },
+          {
+            path: "output_tokens",
+            dest: "output_tokens",
+            required: !0
+          }
+        ]
+      },
+      {
+        api_flavor: "responses",
+        root: "usage",
+        model_path: "model",
+        mappings: [
+          {
+            path: "input_tokens",
+            dest: "input_tokens",
+            required: !0
+          },
+          {
+            path: ["input_tokens_details", "cached_tokens"],
+            dest: "cache_read_tokens",
+            required: !1
+          },
+          {
+            path: "output_tokens",
+            dest: "output_tokens",
+            required: !0
+          }
+        ]
+      },
       {
         api_flavor: "chat",
         root: "usage",
@@ -9106,6 +9373,11 @@ Compared to other leading proprietary and open-weights models Command A delivers
             dest: "input_tokens",
             required: !0
           },
+          {
+            path: ["prompt_tokens_details", "cached_tokens"],
+            dest: "cache_read_tokens",
+            required: !1
+          },
           {
             path: "completion_tokens",
             dest: "output_tokens",
@@ -12572,6 +12844,43 @@ Compared to other leading proprietary and open-weights models Command A delivers
           output_mtok: 15
         }
       },
+      {
+        id: "anthropic/claude-sonnet-5",
+        match: {
+          or: [
+            {
+              equals: "anthropic/claude-sonnet-5"
+            },
+            {
+              equals: "anthropic/claude-sonnet-5:beta"
+            }
+          ]
+        },
+        context_window: 1e6,
+        price_comments: "Flat pricing across full 1M context window (no tiered pricing). Introductory pricing ($2/$10 per MTok) applies through 2026-08-31; standard ($3/$15) from 2026-09-01. OpenRouter mirrors Anthropic first-party pricing; $2/$10 verified live via the OpenRouter API on 2026-06-30. Ref: https://openrouter.ai/anthropic/claude-sonnet-5",
+        prices: [
+          {
+            prices: {
+              input_mtok: 2,
+              cache_write_mtok: 2.5,
+              cache_read_mtok: 0.2,
+              output_mtok: 10
+            }
+          },
+          {
+            constraint: {
+              start_date: "2026-09-01",
+              type: "start_date"
+            },
+            prices: {
+              input_mtok: 3,
+              cache_write_mtok: 3.75,
+              cache_read_mtok: 0.3,
+              output_mtok: 15
+            }
+          }
+        ]
+      },
       {
         id: "anubis-pro-105b-v1",
         name: "Anubis Pro 105B V1",
@@ -14132,7 +14441,14 @@ Compared to other leading proprietary and open-weights models Command A delivers
         id: "google/gemini-3.5-flash",
         name: "Gemini 3.5 Flash",
         match: {
-          equals: "google/gemini-3.5-flash"
+          or: [
+            {
+              equals: "google/gemini-3.5-flash"
+            },
+            {
+              regex: "^google/gemini-3\\.5-flash-\\d{8}$"
+            }
+          ]
         },
         prices: {
           input_mtok: 1.5,
@@ -19298,7 +19614,14 @@ Compared to other leading proprietary and open-weights models Command A delivers
         id: "x-ai/grok-4.3",
         name: "Grok 4.3",
         match: {
-          equals: "x-ai/grok-4.3"
+          or: [
+            {
+              equals: "x-ai/grok-4.3"
+            },
+            {
+              regex: "^x-ai/grok-4\\.3-\\d{8}$"
+            }
+          ]
         },
         prices: {
           input_mtok: 1.25,
@@ -21505,6 +21828,15 @@ Compared to other leading proprietary and open-weights models Command A delivers
             {
               equals: "grok-4.3"
             },
+            {
+              regex: "^grok-4\\.3-\\d{8}$"
+            },
+            {
+              equals: "x-ai/grok-4.3"
+            },
+            {
+              regex: "^x-ai/grok-4\\.3-\\d{8}$"
+            },
             {
               equals: "grok-4.3-latest"
             },
@@ -22012,7 +22344,7 @@ function j(e, t, i) {
 function H(e) {
   return M?.(), x(k, e);
 }
-function U(e, t, i) {
+function $(e, t, i) {
   if (i = i ?? "default", !e.extractors)
     throw new Error("No extraction logic defined for this provider");
   const a = e.extractors.find((s) => s.api_flavor === i);
@@ -22105,7 +22437,7 @@ const q = {
   guard: (e) => typeof e == "number",
   name: "number"
 }, h = (e, t) => [...e.map(L), ...t.map(L)].join("."), L = (e) => typeof e == "string" ? e : JSON.stringify(e);
-class $ {
+class U {
   base;
   tiers;
   constructor(t) {
@@ -22114,9 +22446,9 @@ class $ {
 }
 export {
   R as REMOTE_DATA_JSON_URL,
-  $ as TieredPrices,
+  U as TieredPrices,
   j as calcPrice,
-  U as extractUsage,
+  $ as extractUsage,
   H as findProvider,
   O as updatePrices,
   K as waitForUpdate

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pydantic/genai-prices",
-  "version": "0.0.67",
+  "version": "0.0.69",
   "description": "Calculate prices for calling LLM inference APIs",
   "author": "Pydantic Team",
   "type": "module",