pi-fireworks-provider 1.5.1 → 1.5.3

package/README.md

@@ -15,7 +15,7 @@ _Kimi, MiniMax, GLM, DeepSeek, GPT-OSS — via Fireworks AI's Anthropic Messages
 
 ## Features
 
-- **39+ AI Models** including Kimi K2.5, MiniMax M2.5, GLM 4.5/4.7/5, DeepSeek V3.1/V3.2, DeepSeek V4 Flash, and GPT-OSS
+- **43+ AI Models** including Kimi K2.5, MiniMax M2.5, GLM 4.5/4.7/5, DeepSeek V3.1/V3.2, DeepSeek V4 Flash, and GPT-OSS
 - **Dual API support** via Fireworks AI's Anthropic Messages and OpenAI-compatible completions endpoints (per-model routing, matching pi core's Fireworks provider)
 - **Service tiers** — toggle Fireworks `priority` vs `standard` per request on supported models (with priority pricing reflected in cost tracking), via a keybinding, `/fireworks-tier`, and a footer status area
 - **Preserved thinking** — toggle Fireworks' `reasoning_history: "preserved"` so prior assistant reasoning is retained across turns (better multi-turn recall; uses more tokens), via the `/fireworks-settings` panel, with a model-select notification. Matches neuralwatt/makora's settings-only UX, adapted to Fireworks' single global `reasoning_history` knob
@@ -75,10 +75,11 @@ pi
 | DeepSeek V3.1 | Text | 164K | 164K | $0.56 | $1.68 |
 | DeepSeek V3.2 | Text | 164K | 160K | $0.56 | $1.68 |
 | DeepSeek V4 Flash | Text | 1.0M | 384K | $0.14 | $0.28 |
+| DeepSeek V4 Flash 0731 | Text | 1.0M | 384K | $0.14 | $0.28 |
 | DeepSeek V4 Pro | Text | 1.0M | 384K | $1.74 | $3.48 |
-| DeepSeek V4 Pro (router) | Text | 1.0M | 1.0M | $1.74 | $3.48 |
-| Gemma 4 26B A4B IT | Text + Image | 262K | 0 | Free | Free |
-| Gemma 4 31B IT | Text + Image | 262K | 0 | Free | Free |
+| DeepSeek V4 Pro (router) | Text | 1.0M | 384K | $1.74 | $3.48 |
+| Gemma 4 26B A4B IT | Text + Image | 262K | 33K | Free | Free |
+| Gemma 4 31B IT | Text + Image | 262K | 33K | Free | Free |
 | GLM 4.5 | Text | 131K | 131K | $0.55 | $2.19 |
 | GLM 4.5 Air | Text | 131K | 131K | $0.22 | $0.88 |
 | GLM 4.7 | Text | 203K | 198K | $0.60 | $2.20 |
@@ -90,6 +91,7 @@ pi
 | GLM 5.2 Fast | Text | 1.0M | 131K | $2.10 | $6.60 |
 | GPT OSS 120B | Text | 131K | 33K | $0.15 | $0.60 |
 | GPT OSS 20B | Text | 131K | 33K | $0.07 | $0.30 |
+| Inkling | Text + Image | 1.0M | 262K | $1.00 | $4.05 |
 | Kimi K2 Instruct | Text | 131K | 16K | $1.00 | $3.00 |
 | Kimi K2 Thinking | Text | 262K | 256K | $0.60 | $2.50 |
 | Kimi K2.5 | Text + Image | 262K | 256K | $0.60 | $3.00 |
@@ -100,17 +102,19 @@ pi
 | Kimi K2.6 Turbo | Text + Image | 262K | 262K | $2.00 | $8.00 |
 | Kimi K2.7 Code | Text + Image | 262K | 262K | $0.95 | $4.00 |
 | Kimi K2.7 Code Fast | Text + Image | 262K | 262K | $1.90 | $8.00 |
-| Llama 3.3 70B Instruct | Text | 131K | 0 | Free | Free |
-| MiniMax M2.7 (router) | Text | 204K | 0 | $0.30 | $1.20 |
+| Kimi K3 | Text + Image | 1.0M | 131K | $3.00 | $15.00 |
+| Kimi K3 Fast | Text + Image | 1.0M | 131K | $4.50 | $22.50 |
+| Llama 3.3 70B Instruct | Text | 131K | 4K | $0.90 | $0.90 |
+| MiniMax M2.7 (router) | Text | 204K | 131K | $0.30 | $1.20 |
 | MiniMax-M2.1 | Text | 197K | 200K | $0.30 | $1.20 |
 | MiniMax-M2.5 | Text | 197K | 197K | $0.30 | $1.20 |
 | MiniMax-M2.7 | Text | 197K | 197K | $0.30 | $1.20 |
-| MiniMax-M3 | Text | 512K | 512K | $0.30 | $1.20 |
-| NVIDIA Nemotron 3 Ultra NVFP4 | Text | 262K | 0 | Free | Free |
+| MiniMax-M3 | Text + Image | 512K | 512K | $0.30 | $1.20 |
+| NVIDIA Nemotron 3 Ultra NVFP4 | Text | 262K | 66K | $0.60 | $2.40 |
 | Qwen 3.7 Plus | Text + Image | 262K | 66K | $0.40 | $1.60 |
-| Qwen3 8B | Text | 41K | 0 | Free | Free |
-| Qwen3 VL 30B A3B Instruct | Text + Image | 262K | 0 | Free | Free |
-| Qwen3 VL 30B A3B Thinking | Text + Image | 262K | 0 | Free | Free |
+| Qwen3 8B | Text | 41K | 41K | $0.20 | $0.20 |
+| Qwen3 VL 30B A3B Instruct | Text + Image | 262K | 33K | $0.50 | $0.50 |
+| Qwen3 VL 30B A3B Thinking | Text + Image | 262K | 33K | $0.50 | $0.50 |
 *Costs are per million tokens. Prices subject to change - check [fireworks.ai](https://fireworks.ai) for current pricing.*
 
 ## Service Tiers

package/custom-models.json

@@ -13,7 +13,7 @@
       "text"
     ],
     "contextWindow": 1048576,
-    "maxTokens": 0
+    "maxTokens": 384000
   },
   {
     "id": "accounts/fireworks/routers/kimi-k2p5-fast",
@@ -30,7 +30,7 @@
       "image"
     ],
     "contextWindow": 262144,
-    "maxTokens": 0
+    "maxTokens": 256000
   },
   {
     "id": "accounts/fireworks/routers/kimi-k2p6-turbo",
@@ -47,7 +47,7 @@
       "image"
     ],
     "contextWindow": 262144,
-    "maxTokens": 0
+    "maxTokens": 262000
   },
   {
     "id": "accounts/fireworks/routers/kimi-k2p6",
@@ -96,7 +96,7 @@
       "text"
     ],
     "contextWindow": 204000,
-    "maxTokens": 0
+    "maxTokens": 131072
   },
   {
     "id": "accounts/fireworks/routers/glm-5-fast",
@@ -129,7 +129,7 @@
       "image"
     ],
     "contextWindow": 262000,
-    "maxTokens": 0
+    "maxTokens": 32768
   },
   {
     "id": "accounts/fireworks/models/gemma-4-26b-a4b-it",
@@ -146,7 +146,7 @@
       "image"
     ],
     "contextWindow": 262000,
-    "maxTokens": 0
+    "maxTokens": 32768
   },
   {
     "id": "accounts/fireworks/models/qwen3p7-plus",
@@ -214,5 +214,25 @@
     ],
     "contextWindow": 262000,
     "maxTokens": 262000
+  },
+  {
+    "id": "accounts/fireworks/routers/kimi-k3-fast",
+    "name": "Kimi K3 Fast",
+    "reasoning": true,
+    "input": [
+      "text",
+      "image"
+    ],
+    "cost": {
+      "input": 4.5,
+      "output": 22.5,
+      "cacheRead": 0.45,
+      "cacheWrite": 0
+    },
+    "contextWindow": 1048576,
+    "maxTokens": 131072,
+    "compat": {
+      "supportsReasoningEffort": true
+    }
   }
 ]

package/deprecated-models.json

@@ -0,0 +1 @@
+{}

package/index.ts

@@ -30,6 +30,7 @@ import type { Input, matchesKey, Key, KeyId, truncateToWidth, visibleWidth, wrap
 import modelsData from "./models.json" with { type: "json" };
 import customModelsData from "./custom-models.json" with { type: "json" };
 import patchData from "./patch.json" with { type: "json" };
+import deprecatedData from "./deprecated-models.json" with { type: "json" };
 import fs from "fs";
 import path from "path";
 
@@ -118,7 +119,10 @@ function applyPatch(model: JsonModel, patch: PatchEntry): JsonModel {
 function buildModels(base: JsonModel[], custom: JsonModel[], patch: PatchData): JsonModel[] {
   const modelMap = new Map<string, JsonModel>();
 
-  for (const model of base) {
+  // Seed with the base list plus grace-period deprecated models so patch.json
+  // entries apply to deprecated models exactly as while the model was live
+  // (withDeprecated keeps live data on id conflicts).
+  for (const model of withDeprecated(base)) {
     modelMap.set(model.id, model);
   }
 
@@ -245,6 +249,35 @@ function mergeWithEmbedded(liveModels: JsonModel[], embeddedModels: JsonModel[])
   return result;
 }
 
+// Grace period for delisted models. When the provider API stops listing a
+// model, update-models.js moves its last-known definition into
+// deprecated-models.json (stamped with deprecatedAt) instead of dropping it.
+// For 14 days the model keeps working here so in-flight sessions and saved
+// model settings do not break; afterwards it is evicted permanently.
+const DEPRECATED_MODEL_TTL_MS = 14 * 24 * 60 * 60 * 1000;
+
+// Grace-period deprecated models with deprecation metadata stripped.
+function activeDeprecatedModels(): JsonModel[] {
+  const now = Date.now();
+  const result: JsonModel[] = [];
+  for (const entry of Object.values(deprecatedData as Record<string, JsonModel & { deprecatedAt?: string }>)) {
+    if (!entry?.id) continue;
+    const removedAt = Date.parse(entry.deprecatedAt ?? "");
+    if (Number.isNaN(removedAt) || now - removedAt > DEPRECATED_MODEL_TTL_MS) continue;
+    const model = { ...entry } as JsonModel & { deprecatedAt?: string };
+    delete model.deprecatedAt;
+    result.push(model);
+  }
+  return result;
+}
+
+// Append grace-period deprecated models the list does not already have (live data wins).
+function withDeprecated(models: JsonModel[]): JsonModel[] {
+  const seen = new Set(models.map((m) => m.id));
+  const extras = activeDeprecatedModels().filter((m) => !seen.has(m.id));
+  return extras.length > 0 ? [...models, ...extras] : models;
+}
+
 function loadStaleModels(embeddedModels: JsonModel[]): JsonModel[] {
   const cached = loadCachedModels();
   if (!cached || cached.length === 0) return embeddedModels;

package/models.json

@@ -47,6 +47,22 @@
     "contextWindow": 1048576,
     "maxTokens": 0
   },
+  {
+    "id": "accounts/fireworks/models/deepseek-v4-flash-0731",
+    "name": "DeepSeek-V4-Flash-0731",
+    "reasoning": false,
+    "input": [
+      "text"
+    ],
+    "cost": {
+      "input": 0,
+      "output": 0,
+      "cacheRead": 0,
+      "cacheWrite": 0
+    },
+    "contextWindow": 1048576,
+    "maxTokens": 0
+  },
   {
     "id": "accounts/fireworks/models/deepseek-v4-pro",
     "name": "DeepSeek-V4-Pro",
@@ -191,6 +207,23 @@
     "contextWindow": 131072,
     "maxTokens": 0
   },
+  {
+    "id": "accounts/fireworks/models/inkling",
+    "name": "Inkling",
+    "reasoning": false,
+    "input": [
+      "text",
+      "image"
+    ],
+    "cost": {
+      "input": 0,
+      "output": 0,
+      "cacheRead": 0,
+      "cacheWrite": 0
+    },
+    "contextWindow": 1048576,
+    "maxTokens": 0
+  },
   {
     "id": "accounts/fireworks/models/kimi-k2-instruct",
     "name": "Kimi K2 Instruct",
@@ -274,6 +307,23 @@
     "contextWindow": 262144,
     "maxTokens": 0
   },
+  {
+    "id": "accounts/fireworks/models/kimi-k3",
+    "name": "Kimi K3",
+    "reasoning": false,
+    "input": [
+      "text",
+      "image"
+    ],
+    "cost": {
+      "input": 0,
+      "output": 0,
+      "cacheRead": 0,
+      "cacheWrite": 0
+    },
+    "contextWindow": 1048576,
+    "maxTokens": 0
+  },
   {
     "id": "accounts/fireworks/models/llama-v3p3-70b-instruct",
     "name": "Llama 3.3 70B Instruct",
@@ -343,8 +393,7 @@
     "name": "Minimax M3",
     "reasoning": false,
     "input": [
-      "text",
-      "image"
+      "text"
     ],
     "cost": {
       "input": 0,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-fireworks-provider",
-  "version": "1.5.1",
+  "version": "1.5.3",
   "description": "Fireworks AI provider extension for pi - Access Kimi, MiniMax, GLM, DeepSeek, and GPT-OSS models through the Fireworks AI API",
   "type": "module",
   "main": "index.ts",
@@ -22,6 +22,7 @@
   "files": [
     "index.ts",
     "models.json",
+    "deprecated-models.json",
     "custom-models.json",
     "patch.json",
     "README.md",

package/patch.json

@@ -2,7 +2,9 @@
   "accounts/fireworks/models/deepseek-v3p1": {
     "name": "DeepSeek V3.1",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.56,
       "output": 1.68,
@@ -18,7 +20,9 @@
   "accounts/fireworks/models/deepseek-v3p2": {
     "name": "DeepSeek V3.2",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.56,
       "output": 1.68,
@@ -34,14 +38,9 @@
   "accounts/fireworks/models/deepseek-v4-flash": {
     "name": "DeepSeek V4 Flash",
     "reasoning": true,
-    "thinkingLevelMap": {
-      "off": "none",
-      "minimal": null,
-      "low": "high",
-      "medium": "high",
-      "max": "max"
-    },
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.14,
       "output": 0.28,
@@ -57,14 +56,9 @@
   "accounts/fireworks/models/deepseek-v4-pro": {
     "name": "DeepSeek V4 Pro",
     "reasoning": true,
-    "thinkingLevelMap": {
-      "off": "none",
-      "minimal": null,
-      "low": "high",
-      "medium": "high",
-      "max": "max"
-    },
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 1.74,
       "output": 3.48,
@@ -80,7 +74,9 @@
   "accounts/fireworks/models/glm-4p5": {
     "name": "GLM 4.5",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.55,
       "output": 2.19,
@@ -96,7 +92,9 @@
   "accounts/fireworks/models/glm-4p5-air": {
     "name": "GLM 4.5 Air",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.22,
       "output": 0.88,
@@ -112,7 +110,9 @@
   "accounts/fireworks/models/glm-4p7": {
     "name": "GLM 4.7",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.6,
       "output": 2.2,
@@ -128,7 +128,9 @@
   "accounts/fireworks/models/glm-5": {
     "name": "GLM 5",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 1,
       "output": 3.2,
@@ -144,7 +146,9 @@
   "accounts/fireworks/models/glm-5p1": {
     "name": "GLM 5.1",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 1.4,
       "output": 4.4,
@@ -167,13 +171,17 @@
       "minimal": null,
       "low": "high",
       "medium": "high",
+      "high": "high",
+      "xhigh": null,
       "max": "max"
     },
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 1.4,
       "output": 4.4,
-      "cacheRead": 0.26,
+      "cacheRead": 0.14,
       "cacheWrite": 0
     },
     "contextWindow": 1048575,
@@ -186,7 +194,9 @@
   "accounts/fireworks/models/gpt-oss-120b": {
     "name": "GPT OSS 120B",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.15,
       "output": 0.6,
@@ -202,7 +212,9 @@
   "accounts/fireworks/models/gpt-oss-20b": {
     "name": "GPT OSS 20B",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.07,
       "output": 0.3,
@@ -218,7 +230,10 @@
   "accounts/fireworks/models/gemma-4-26b-a4b-it": {
     "name": "Gemma 4 26B A4B IT",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0,
       "output": 0,
@@ -228,12 +243,16 @@
     "contextWindow": 262000,
     "compat": {
       "supportsReasoningEffort": true
-    }
+    },
+    "maxTokens": 32768
   },
   "accounts/fireworks/models/gemma-4-31b-it": {
     "name": "Gemma 4 31B IT",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0,
       "output": 0,
@@ -243,12 +262,15 @@
     "contextWindow": 262000,
     "compat": {
       "supportsReasoningEffort": true
-    }
+    },
+    "maxTokens": 32768
   },
   "accounts/fireworks/models/kimi-k2-instruct": {
     "name": "Kimi K2 Instruct",
     "reasoning": false,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 1,
       "output": 3,
@@ -261,7 +283,9 @@
   "accounts/fireworks/models/kimi-k2-thinking": {
     "name": "Kimi K2 Thinking",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.6,
       "output": 2.5,
@@ -277,7 +301,10 @@
   "accounts/fireworks/models/kimi-k2p5": {
     "name": "Kimi K2.5",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0.6,
       "output": 3,
@@ -293,7 +320,10 @@
   "accounts/fireworks/models/kimi-k2p6": {
     "name": "Kimi K2.6",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0.95,
       "output": 4,
@@ -309,7 +339,10 @@
   "accounts/fireworks/models/kimi-k2p7-code": {
     "name": "Kimi K2.7 Code",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0.95,
       "output": 4,
@@ -325,7 +358,9 @@
   "accounts/fireworks/models/minimax-m2p1": {
     "name": "MiniMax-M2.1",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.3,
       "output": 1.2,
@@ -341,7 +376,9 @@
   "accounts/fireworks/models/minimax-m2p5": {
     "name": "MiniMax-M2.5",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.3,
       "output": 1.2,
@@ -357,7 +394,9 @@
   "accounts/fireworks/models/minimax-m2p7": {
     "name": "MiniMax-M2.7",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.3,
       "output": 1.2,
@@ -373,7 +412,10 @@
   "accounts/fireworks/models/minimax-m3": {
     "name": "MiniMax-M3",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0.3,
       "output": 1.2,
@@ -389,7 +431,10 @@
   "accounts/fireworks/models/qwen3p7-plus": {
     "name": "Qwen 3.7 Plus",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0.4,
       "output": 1.6,
@@ -412,7 +457,9 @@
       "medium": "high",
       "max": "max"
     },
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 1.74,
       "output": 3.48,
@@ -420,7 +467,7 @@
       "cacheWrite": 0
     },
     "contextWindow": 1048576,
-    "maxTokens": 1048576,
+    "maxTokens": 384000,
     "compat": {
       "supportsReasoningEffort": true
     }
@@ -428,7 +475,9 @@
   "accounts/fireworks/routers/glm-5-fast": {
     "name": "GLM 5 Fast",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 1,
       "output": 3.2,
@@ -444,7 +493,9 @@
   "accounts/fireworks/routers/glm-5p1-fast": {
     "name": "GLM 5.1 Fast",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 2.8,
       "output": 8.8,
@@ -465,9 +516,13 @@
       "minimal": null,
       "low": "high",
       "medium": "high",
+      "high": "high",
+      "xhigh": null,
       "max": "max"
     },
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 2.1,
       "output": 6.6,
@@ -483,7 +538,10 @@
   "accounts/fireworks/routers/kimi-k2p5-fast": {
     "name": "Kimi K2.5 Fast (router)",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0.6,
       "output": 3,
@@ -499,7 +557,10 @@
   "accounts/fireworks/routers/kimi-k2p6": {
     "name": "Kimi K2.6 (router)",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 0.95,
       "output": 4,
@@ -515,7 +576,10 @@
   "accounts/fireworks/routers/kimi-k2p6-fast": {
     "name": "Kimi K2.6 Fast",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 2,
       "output": 8,
@@ -531,7 +595,10 @@
   "accounts/fireworks/routers/kimi-k2p6-turbo": {
     "name": "Kimi K2.6 Turbo",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 2,
       "output": 8,
@@ -547,7 +614,10 @@
   "accounts/fireworks/routers/kimi-k2p7-code-fast": {
     "name": "Kimi K2.7 Code Fast",
     "reasoning": true,
-    "input": ["text", "image"],
+    "input": [
+      "text",
+      "image"
+    ],
     "cost": {
       "input": 1.9,
       "output": 8,
@@ -563,7 +633,9 @@
   "accounts/fireworks/routers/minimax-m2p7": {
     "name": "MiniMax M2.7 (router)",
     "reasoning": true,
-    "input": ["text"],
+    "input": [
+      "text"
+    ],
     "cost": {
       "input": 0.3,
       "output": 1.2,
@@ -571,6 +643,156 @@
       "cacheWrite": 0
     },
     "contextWindow": 204000,
+    "compat": {
+      "supportsReasoningEffort": true
+    },
+    "maxTokens": 131072
+  },
+  "accounts/fireworks/models/deepseek-v4-flash-0731": {
+    "name": "DeepSeek V4 Flash 0731",
+    "reasoning": true,
+    "thinkingLevelMap": {
+      "off": "none",
+      "minimal": null,
+      "low": "high",
+      "medium": "high",
+      "max": "max"
+    },
+    "input": [
+      "text"
+    ],
+    "cost": {
+      "input": 0.14,
+      "output": 0.28,
+      "cacheRead": 0.028,
+      "cacheWrite": 0
+    },
+    "contextWindow": 1000000,
+    "maxTokens": 384000,
+    "compat": {
+      "supportsReasoningEffort": true
+    }
+  },
+  "accounts/fireworks/models/inkling": {
+    "name": "Inkling",
+    "reasoning": true,
+    "input": [
+      "text",
+      "image"
+    ],
+    "cost": {
+      "input": 1,
+      "output": 4.05,
+      "cacheRead": 0.17,
+      "cacheWrite": 0
+    },
+    "contextWindow": 1048576,
+    "maxTokens": 262144,
+    "compat": {
+      "supportsReasoningEffort": true
+    }
+  },
+  "accounts/fireworks/models/kimi-k3": {
+    "name": "Kimi K3",
+    "reasoning": true,
+    "input": [
+      "text",
+      "image"
+    ],
+    "cost": {
+      "input": 3,
+      "output": 15,
+      "cacheRead": 0.3,
+      "cacheWrite": 0
+    },
+    "contextWindow": 1048576,
+    "maxTokens": 131072,
+    "compat": {
+      "supportsReasoningEffort": true
+    }
+  },
+  "accounts/fireworks/models/llama-v3p3-70b-instruct": {
+    "name": "Llama 3.3 70B Instruct",
+    "reasoning": false,
+    "input": [
+      "text"
+    ],
+    "cost": {
+      "input": 0.9,
+      "output": 0.9,
+      "cacheRead": 0,
+      "cacheWrite": 0
+    },
+    "contextWindow": 131072,
+    "maxTokens": 4096
+  },
+  "accounts/fireworks/models/nemotron-3-ultra-nvfp4": {
+    "name": "NVIDIA Nemotron 3 Ultra NVFP4",
+    "reasoning": true,
+    "input": [
+      "text"
+    ],
+    "cost": {
+      "input": 0.6,
+      "output": 2.4,
+      "cacheRead": 0.12,
+      "cacheWrite": 0
+    },
+    "contextWindow": 262144,
+    "maxTokens": 65536,
+    "compat": {
+      "supportsReasoningEffort": true
+    }
+  },
+  "accounts/fireworks/models/qwen3-8b": {
+    "name": "Qwen3 8B",
+    "reasoning": true,
+    "input": [
+      "text"
+    ],
+    "cost": {
+      "input": 0.2,
+      "output": 0.2,
+      "cacheRead": 0,
+      "cacheWrite": 0
+    },
+    "contextWindow": 40960,
+    "maxTokens": 40960,
+    "compat": {
+      "supportsReasoningEffort": true
+    }
+  },
+  "accounts/fireworks/models/qwen3-vl-30b-a3b-instruct": {
+    "name": "Qwen3 VL 30B A3B Instruct",
+    "reasoning": false,
+    "input": [
+      "text",
+      "image"
+    ],
+    "cost": {
+      "input": 0.5,
+      "output": 0.5,
+      "cacheRead": 0,
+      "cacheWrite": 0
+    },
+    "contextWindow": 262144,
+    "maxTokens": 32768
+  },
+  "accounts/fireworks/models/qwen3-vl-30b-a3b-thinking": {
+    "name": "Qwen3 VL 30B A3B Thinking",
+    "reasoning": true,
+    "input": [
+      "text",
+      "image"
+    ],
+    "cost": {
+      "input": 0.5,
+      "output": 0.5,
+      "cacheRead": 0,
+      "cacheWrite": 0
+    },
+    "contextWindow": 262144,
+    "maxTokens": 32768,
     "compat": {
       "supportsReasoningEffort": true
     }