ton/pi_harness
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(ai): show implied Kimi Coding subscription costs

Browse Source
This commit is contained in:
Armin Ronacher
2026-07-17 10:00:40 +02:00
parent 58575888f5
commit 8881e17625
7 changed files with 70 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files
packages/ai/CHANGELOG.md
+4
View File
@@ -2,6 +2,10 @@
## [Unreleased]
### Fixed
- Fixed Kimi Coding subscription models to report API-equivalent implied costs when models.dev reports zero pricing.
## [0.80.10] - 2026-07-16
### Fixed
packages/ai/scripts/generate-models.ts
+14 -4
View File
@@ -241,6 +241,15 @@ const KIMI_K3_COST = {
cacheRead: 0.3,
cacheWrite: 0,
} as const;
// Kimi Coding is subscription-backed, so models.dev reports zero cost. Use the
// equivalent Moonshot API rates to estimate the value of subscription usage.
const KIMI_CODING_IMPLIED_COSTS: Record<string, Model<Api>["cost"]> = {
k2p7: { input: 0.95, output: 4, cacheRead: 0.19, cacheWrite: 0 },
k3: KIMI_K3_COST,
"kimi-for-coding": { input: 0.95, output: 4, cacheRead: 0.19, cacheWrite: 0 },
"kimi-for-coding-highspeed": { input: 1.9, output: 8, cacheRead: 0.38, cacheWrite: 0 },
"kimi-k2-thinking": { input: 0.6, output: 2.5, cacheRead: 0.15, cacheWrite: 0 },
};
const OPENROUTER_KIMI_K3_MODEL_IDS = new Set(["moonshotai/kimi-k3", "~moonshotai/kimi-latest"]);
const ANT_LING_RING_THINKING_LEVEL_MAP = {
@@ -1642,6 +1651,7 @@ async function loadModelsDevData(): Promise<Model<any>[]> {
const normalizedName = kimiAliases.has(modelId) ? "Kimi For Coding" : m.name || normalizedId;
const isKimiK3 = normalizedId === "k3";
const allowEmptySignature = isKimiK3 || normalizedId === "kimi-for-coding";
const impliedCost = KIMI_CODING_IMPLIED_COSTS[normalizedId];
models.push({
id: normalizedId,
@@ -1659,10 +1669,10 @@ async function loadModelsDevData(): Promise<Model<any>[]> {
...(isKimiK3 ? { thinkingLevelMap: KIMI_K3_THINKING_LEVEL_MAP } : {}),
input: m.modalities?.input?.includes("image") ? ["text", "image"] : ["text"],
cost: {
input: m.cost?.input || 0,
output: m.cost?.output || 0,
cacheRead: m.cost?.cache_read || 0,
cacheWrite: m.cost?.cache_write || 0,
input: m.cost?.input || impliedCost?.input || 0,
output: m.cost?.output || impliedCost?.output || 0,
cacheRead: m.cost?.cache_read || impliedCost?.cacheRead || 0,
cacheWrite: m.cost?.cache_write || impliedCost?.cacheWrite || 0,
},
contextWindow: m.limit?.context || 4096,
maxTokens: m.limit?.output || 4096,
packages/ai/src/providers/kimi-coding.models.ts
+15 -15
View File
@@ -15,9 +15,9 @@ export const KIMI_CODING_MODELS = {
reasoning: true,
input: ["text", "image"],
cost: {
input: 0,
output: 0,
cacheRead: 0,
input: 0.95,
output: 4,
cacheRead: 0.19,
cacheWrite: 0,
},
contextWindow: 262144,
@@ -35,9 +35,9 @@ export const KIMI_CODING_MODELS = {
thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
input: ["text", "image"],
cost: {
input: 0,
output: 0,
cacheRead: 0,
input: 3,
output: 15,
cacheRead: 0.3,
cacheWrite: 0,
},
contextWindow: 1048576,
@@ -54,9 +54,9 @@ export const KIMI_CODING_MODELS = {
reasoning: true,
input: ["text", "image"],
cost: {
input: 0,
output: 0,
cacheRead: 0,
input: 0.95,
output: 4,
cacheRead: 0.19,
cacheWrite: 0,
},
contextWindow: 262144,
@@ -73,9 +73,9 @@ export const KIMI_CODING_MODELS = {
reasoning: true,
input: ["text", "image"],
cost: {
input: 0,
output: 0,
cacheRead: 0,
input: 1.9,
output: 8,
cacheRead: 0.38,
cacheWrite: 0,
},
contextWindow: 262144,
@@ -92,9 +92,9 @@ export const KIMI_CODING_MODELS = {
reasoning: true,
input: ["text"],
cost: {
input: 0,
output: 0,
cacheRead: 0,
input: 0.6,
output: 2.5,
cacheRead: 0.15,
cacheWrite: 0,
},
contextWindow: 262144,
packages/ai/test/providers.test.ts
+15
View File
@@ -56,6 +56,21 @@ describe("builtin providers", () => {
}
});
it("uses API-equivalent implied pricing for Kimi Coding subscription models", () => {
const models = builtinModels();
const expectedCosts = {
k2p7: { input: 0.95, output: 4, cacheRead: 0.19, cacheWrite: 0 },
k3: { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 0 },
"kimi-for-coding": { input: 0.95, output: 4, cacheRead: 0.19, cacheWrite: 0 },
"kimi-for-coding-highspeed": { input: 1.9, output: 8, cacheRead: 0.38, cacheWrite: 0 },
"kimi-k2-thinking": { input: 0.6, output: 2.5, cacheRead: 0.15, cacheWrite: 0 },
};
for (const [modelId, cost] of Object.entries(expectedCosts)) {
expect(models.getModel("kimi-coding", modelId)?.cost).toEqual(cost);
}
});
it("resolves anthropic auth from env with OAuth token precedence", async () => {
const models = createModels({
authContext: fakeAuthContext({ ANTHROPIC_API_KEY: "key", ANTHROPIC_OAUTH_TOKEN: "oauth-token" }),