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

fix(ai): add GPT-5.4 and GPT-5.5 long-context pricing

Browse Source
This commit is contained in:
Armin Ronacher
2026-07-09 23:13:58 +02:00
parent a9ecf301fb
commit 3664806f2f
4 changed files with 31 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files
packages/ai/CHANGELOG.md
+1
View File
@@ -10,6 +10,7 @@
### Fixed
- Fixed post-compaction output-token budgeting to ignore stale assistant usage from before the compaction boundary ([#6464](https://github.com/earendil-works/pi/issues/6464)).
- Fixed GPT-5.4 and GPT-5.5 long-context cost accounting while retaining the intentional 272K default context limit for models that require an explicit override.
- Fixed GPT-5.6 metadata to keep direct OpenAI requests in the 272K short-context tier while exposing the Codex backend's 372K context window with long-context pricing.
## [0.80.5] - 2026-07-09
packages/ai/scripts/generate-models.ts
+24 -11
View File
@@ -195,7 +195,22 @@ const ANT_LING_RING_THINKING_LEVEL_MAP = {
const MODELS_DEV_OPENAI_UNSUPPORTED_MODEL_IDS = new Set(["gpt-5.6"]);
const OPENAI_LONG_CONTEXT_INPUT_THRESHOLD = 272000;
const OPENAI_GPT_56_MODEL_IDS = new Set(["gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"]);
const OPENAI_SHORT_CONTEXT_CAPPED_MODEL_IDS = new Set([
"gpt-5.4",
"gpt-5.5",
"gpt-5.6-sol",
"gpt-5.6-terra",
"gpt-5.6-luna",
]);
const OPENAI_LONG_CONTEXT_PRICING_MODEL_IDS = new Set([
"gpt-5.4",
"gpt-5.4-pro",
"gpt-5.5",
"gpt-5.5-pro",
"gpt-5.6-sol",
"gpt-5.6-terra",
"gpt-5.6-luna",
]);
function withOpenAiLongContextPricing(cost: Model<Api>["cost"]): Model<Api>["cost"] {
return {
@@ -1695,16 +1710,14 @@ async function generateModels() {
candidate.contextWindow = 272000;
candidate.maxTokens = 128000;
}
// Keep direct OpenAI requests in the short-context pricing tier.
if (
candidate.provider === "openai" &&
(candidate.id === "gpt-5.4" || candidate.id === "gpt-5.5" || OPENAI_GPT_56_MODEL_IDS.has(candidate.id))
) {
// Keep direct OpenAI requests in the short-context pricing tier by default. Users can opt into the
// larger context through model overrides, so retain long-context cost metadata on the capped models.
if (candidate.provider === "openai" && OPENAI_SHORT_CONTEXT_CAPPED_MODEL_IDS.has(candidate.id)) {
candidate.contextWindow = OPENAI_LONG_CONTEXT_INPUT_THRESHOLD;
candidate.maxTokens = 128000;
if (OPENAI_GPT_56_MODEL_IDS.has(candidate.id)) {
candidate.cost = withOpenAiLongContextPricing(candidate.cost);
}
}
if (candidate.provider === "openai" && OPENAI_LONG_CONTEXT_PRICING_MODEL_IDS.has(candidate.id)) {
candidate.cost = withOpenAiLongContextPricing(candidate.cost);
}
// models.dev reports gpt-5-pro output as 272000 (a duplicate of the input sub-limit);
// the actual max output is 128000. Also propagates to the derived Azure clone.
@@ -1950,7 +1963,7 @@ async function generateModels() {
baseUrl: CODEX_BASE_URL,
reasoning: true,
input: ["text", "image"],
cost: { input: 2.5, output: 15, cacheRead: 0.25, cacheWrite: 0 },
cost: withOpenAiLongContextPricing({ input: 2.5, output: 15, cacheRead: 0.25, cacheWrite: 0 }),
contextWindow: CODEX_CONTEXT,
maxTokens: CODEX_MAX_TOKENS,
},
@@ -1974,7 +1987,7 @@ async function generateModels() {
baseUrl: CODEX_BASE_URL,
reasoning: true,
input: ["text", "image"],
cost: { input: 5, output: 30, cacheRead: 0.5, cacheWrite: 0 },
cost: withOpenAiLongContextPricing({ input: 5, output: 30, cacheRead: 0.5, cacheWrite: 0 }),
contextWindow: CODEX_CONTEXT,
maxTokens: CODEX_MAX_TOKENS,
},
packages/ai/src/providers/openai-codex.models.ts
+2
View File
@@ -36,6 +36,7 @@ export const OPENAI_CODEX_MODELS = {
output: 15,
cacheRead: 0.25,
cacheWrite: 0,
tiers: [{"inputTokensAbove":272000,"input":5,"output":22.5,"cacheRead":0.5,"cacheWrite":0}],
},
contextWindow: 272000,
maxTokens: 128000,
@@ -72,6 +73,7 @@ export const OPENAI_CODEX_MODELS = {
output: 30,
cacheRead: 0.5,
cacheWrite: 0,
tiers: [{"inputTokensAbove":272000,"input":10,"output":45,"cacheRead":1,"cacheWrite":0}],
},
contextWindow: 272000,
maxTokens: 128000,
packages/ai/src/providers/openai.models.ts
+4
View File
@@ -512,6 +512,7 @@ export const OPENAI_MODELS = {
output: 15,
cacheRead: 0.25,
cacheWrite: 0,
tiers: [{"inputTokensAbove":272000,"input":5,"output":22.5,"cacheRead":0.5,"cacheWrite":0}],
},
contextWindow: 272000,
maxTokens: 128000,
@@ -566,6 +567,7 @@ export const OPENAI_MODELS = {
output: 180,
cacheRead: 0,
cacheWrite: 0,
tiers: [{"inputTokensAbove":272000,"input":60,"output":270,"cacheRead":0,"cacheWrite":0}],
},
contextWindow: 1050000,
maxTokens: 128000,
@@ -584,6 +586,7 @@ export const OPENAI_MODELS = {
output: 30,
cacheRead: 0.5,
cacheWrite: 0,
tiers: [{"inputTokensAbove":272000,"input":10,"output":45,"cacheRead":1,"cacheWrite":0}],
},
contextWindow: 272000,
maxTokens: 128000,
@@ -602,6 +605,7 @@ export const OPENAI_MODELS = {
output: 180,
cacheRead: 0,
cacheWrite: 0,
tiers: [{"inputTokensAbove":272000,"input":60,"output":270,"cacheRead":0,"cacheWrite":0}],
},
contextWindow: 1050000,
maxTokens: 128000,