Fimeg/letta-code
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

fix(models): recommend GPT-5.3 Codex over GPT-5.2 (#1132)

Browse Source
This commit is contained in:
Sarah Wooders
2026-02-25 11:07:10 -08:00
committed by GitHub
parent 97ed54ad22
commit 4e444429ba
2 changed files with 85 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

69
src/models.json
View File

@@ -226,7 +226,7 @@
"id": "gpt-5.3-codex-plus-pro-high",
"handle": "chatgpt-plus-pro/gpt-5.3-codex",
"label": "GPT-5.3 Codex",
"description": "GPT-5.3 Codex (high reasoning) via ChatGPT Plus/Pro",
"description": "OpenAI's best coding model (high reasoning)",
"isFeatured": true,
"updateArgs": {
"reasoning_effort": "high",
@@ -410,7 +410,6 @@
"handle": "openai/gpt-5.2",
"label": "GPT-5.2",
"description": "Latest general-purpose GPT (med reasoning)",
"isFeatured": true,
"updateArgs": {
"reasoning_effort": "medium",
"verbosity": "medium",
@@ -445,6 +444,72 @@
"parallel_tool_calls": true
}
},
{
"id": "gpt-5.3-codex-none",
"handle": "openai/gpt-5.3-codex",
"label": "GPT-5.3-Codex",
"description": "GPT-5.3 variant (no reasoning) optimized for coding",
"updateArgs": {
"reasoning_effort": "none",
"verbosity": "low",
"context_window": 272000,
"max_output_tokens": 128000,
"parallel_tool_calls": true
}
},
{
"id": "gpt-5.3-codex-low",
"handle": "openai/gpt-5.3-codex",
"label": "GPT-5.3-Codex",
"description": "GPT-5.3 variant (low reasoning) optimized for coding",
"updateArgs": {
"reasoning_effort": "low",
"verbosity": "low",
"context_window": 272000,
"max_output_tokens": 128000,
"parallel_tool_calls": true
}
},
{
"id": "gpt-5.3-codex-medium",
"handle": "openai/gpt-5.3-codex",
"label": "GPT-5.3-Codex",
"description": "GPT-5.3 variant (med reasoning) optimized for coding",
"updateArgs": {
"reasoning_effort": "medium",
"verbosity": "low",
"context_window": 272000,
"max_output_tokens": 128000,
"parallel_tool_calls": true
}
},
{
"id": "gpt-5.3-codex-high",
"handle": "openai/gpt-5.3-codex",
"label": "GPT-5.3-Codex",
"description": "OpenAI's best coding model (high reasoning)",
"isFeatured": true,
"updateArgs": {
"reasoning_effort": "high",
"verbosity": "low",
"context_window": 272000,
"max_output_tokens": 128000,
"parallel_tool_calls": true
}
},
{
"id": "gpt-5.3-codex-xhigh",
"handle": "openai/gpt-5.3-codex",
"label": "GPT-5.3-Codex",
"description": "GPT-5.3 variant (max reasoning) optimized for coding",
"updateArgs": {
"reasoning_effort": "xhigh",
"verbosity": "low",
"context_window": 272000,
"max_output_tokens": 128000,
"parallel_tool_calls": true
}
},
{
"id": "gpt-5.2-codex-none",
"handle": "openai/gpt-5.2-codex",

18
src/tests/model-tier-selection.test.ts
View File

@@ -48,6 +48,24 @@ describe("getReasoningTierOptionsForHandle", () => {
]);
});
test("returns ordered reasoning options for gpt-5.3-codex", () => {
const options = getReasoningTierOptionsForHandle("openai/gpt-5.3-codex");
expect(options.map((option) => option.effort)).toEqual([
"none",
"low",
"medium",
"high",
"xhigh",
]);
expect(options.map((option) => option.modelId)).toEqual([
"gpt-5.3-codex-none",
"gpt-5.3-codex-low",
"gpt-5.3-codex-medium",
"gpt-5.3-codex-high",
"gpt-5.3-codex-xhigh",
]);
});
test("returns byok reasoning options for chatgpt-plus-pro gpt-5.3-codex", () => {
const options = getReasoningTierOptionsForHandle(
"chatgpt-plus-pro/gpt-5.3-codex",