sscoble/pi-mono
1
0
Fork 0
mirror of https://github.com/badlogic/pi-mono.git synced 2026-11-05 23:37:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

fix(ai): pin Kimi K3 Moonshot pricing and expect Haiku 5.5 adaptive thinking

Browse source 
models.dev now lists the input rate as Kimi K3's cache write cost on Moonshot, which does not bill cache writes. Claude Haiku 5.5 uses adaptive thinking, so the metadata test now expects it.
This commit is contained in:
Armin Ronacher 2026-10-07 22:41:41 +02:00
parent dce4ae6f79
commit a2eef9eb60
2 changed files with 13 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

15
packages/ai/scripts/generate-models.ts
View file

@ -2433,12 +2433,15 @@ async function loadModelsDevData(): Promise<Model<any>[]> {
baseUrl,
reasoning: isKimiK3 || m.reasoning === true,
input: m.modalities?.input?.includes("image") ? ["text", "image"] : ["text"],
cost: {
input: m.cost?.input || (isKimiK3 ? KIMI_K3_COST.input : 0),
output: m.cost?.output || (isKimiK3 ? KIMI_K3_COST.output : 0),
cacheRead: m.cost?.cache_read || (isKimiK3 ? KIMI_K3_COST.cacheRead : 0),
cacheWrite: m.cost?.cache_write || (isKimiK3 ? KIMI_K3_COST.cacheWrite : 0),
},
// Moonshot does not bill cache writes; models.dev lists the input rate as cache_write for Kimi K3.
cost: isKimiK3
? { ...KIMI_K3_COST }
: {
input: m.cost?.input || 0,
output: m.cost?.output || 0,
cacheRead: m.cost?.cache_read || 0,
cacheWrite: m.cost?.cache_write || 0,
},
contextWindow: m.limit?.context || 4096,
maxTokens: m.limit?.output || 4096,
compat,