[model-inventory] Model alias inventory update - 2026-09-29
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 2/5
- 预计耗时
- 1-3 小时
- 新手友好度
- 78/100
- Issue 类型
- 功能
- 描述清晰度
- 描述清楚
- 活跃度
- 活跃
- 技术栈
- github-actions, go, javascript
调研方向
首先运行 make refresh-models-json，并将生成的数据与 pkg/cli/data/models.json 和 actions/setup/js/models.json 进行对照检查。添加 issue 中描述的八个缺失模型条目，保留 provider_type 和 wire_api，然后确认两个文件仍然完全一致，并且模型刷新检查通过。
由索引模型根据 Issue 内容生成。
描述
Summary
Pricing gaps found; no alias changes required. All live models are already matched by existing alias globs (sonnet, opus, gpt-6, gpt-5.6, grok, mai-code, gemini-*, etc.).
- Providers queried: OpenAI, Anthropic, Gemini, Copilot SDK, Copilot (reflect)
- Proposed alias changes: 0
- Pricing gaps found: 8 model IDs missing from
models.json - Copilot docs pricing table (Playwright, Step 2.5): skipped this run; docs multiplier cross-check not performed. Pricing was validated against SDK
tokenPricesonly.
Provider Model Counts
| Provider | Models Available | Status |
|---|---|---|
| openai | 135 | ✅ ok |
| anthropic | 11 | ✅ ok |
| gemini | 50 | ✅ ok |
| copilot-sdk | 24 | ✅ ok |
| copilot (reflect) | 42 | ✅ ok (includes embeddings, deprecated 4o/4.1, trajectory-compaction) |
Raw API Fields Discovered
- Copilot SDK:
billing.tokenPrices(input/output/cache/cacheWrite prices,contextMax,maxPromptTokens,longContexttier) andbilling.discountPercent(forauto). - Gemini:
inputTokenLimit/outputTokenLimit, supported generation methods. - OpenAI/Anthropic: id,
owned_by/display_name, created timestamps.
models.json Pricing Analysis
Missing from models.json
| Model ID | Provider | Inferred Pricing (per 1M tokens: in/out/cache-read/cache-write) | Basis |
|---|---|---|---|
| claude-opus-5.5 | copilot (github-copilot) | $4 / $20 / $0.20 / $5 | Copilot SDK tokenPrices (400/2000/20/500 cents) |
| claude-sonnet-5.5 | copilot | $2 / $10 / $0.20 / $2.50 | Copilot SDK |
| gpt-6.1-sol | copilot / openai | $2 / $10 / $0.10 / $2.50 | Copilot SDK |
| gpt-6-sol | copilot / openai | $2 / $10 / $0.20 / $2.50 | Copilot SDK |
| gpt-6-luna | copilot / openai | $0.10 / $0.50 / $0.01 / $0.125 | Copilot SDK |
| claude-opus-5-5 | anthropic | same as claude-opus-5.5 | Anthropic API id |
| claude-sonnet-5-5 | anthropic | same as claude-sonnet-5.5 | Anthropic API id |
| gpt-6-luna, gpt-6-sol, gpt-6.1-sol | openai | same as Copilot entries | OpenAI API id (no openai entries) |
Gemini live IDs (e.g. gemini-2.5-*, gemini-3.1-*, gemma-4-*, veo-*, lyria-*) have no google provider block in models.json; only a subset are listed under github-copilot. No pricing is available from the API for these, so no values are inferred.
auto (SDK discountPercent: 10) is a routing pseudo-model and is not a pricing entry.
Historical entries not currently returned
Retained as historical: claude-fable-5, claude-fable-5.1, claude-opus-4.5, claude-opus-4.6, claude-sonnet-4, claude-sonnet-4.5, claude-sonnet-4.6, gpt-5.1/gpt-5.2*, gpt-5.4-nano, kimi-k2.7-code, kimi-k3, raptor-mini, mai-code-1-flash-picker, and others not in SDK/reflect lists. No removals proposed.
Inferred vs stored pricing discrepancies
None detected for models present in both SDK data and models.json (spot-checked gpt-5.6-sol: stored $4/$20/$0.40/$5 matches SDK; claude-opus-5 matches). Deprecated IDs (gpt-4o-mini, gpt-4.1, gpt-4o, gpt-5.4-nano) intentionally ignored.
Proposed Alias Updates
None. Notes on coverage:
sonnet/opus/sonnet-6xcoverclaude-sonnet-5.5andclaude-opus-5.5.gpt-6coversgpt-6-sol,gpt-6-luna,gpt-6-astra,gpt-6.1-sol;gpt-5.6coversgpt-5.6-luna-utility.grok,mai-code,gemini-3.x-flash,gemini-omni,antigravity,veo,lyria,gemma,deep-research,computer-use,roboticsall match live IDs.- Uncovered but low value:
trajectory-compaction(internal Copilot model),gpt-41-copilot(deprecated 4.1 family). - Task-oriented aliases (
coding,reasoning,vision,summarization) already exist.
Next actions
Run make refresh-models-json, then add the missing entries above to both pkg/cli/data/models.json and actions/setup/js/models.json (kept identical), preserving provider_type/wire_api (responses for OpenAI-vendor Copilot models, anthropic for Claude).
Generated by 📦 Daily Model Inventory Checker · copilot · auto · 21.9 AIC · ⌖ 7.2 AIC · ⊞ 13.6K · ◷
- expires on Oct 6, 2026, 3:59 PM UTC-08:00
- 主要语言
- Go
- 星标
- 5.4k
- 派生
- 596
- 平均合并
- 3 小时 37 分钟
- 30 天内合并 PR
- 833
环境准备
在浏览器里用你自己的 GitHub 账号启动这个项目的开发容器。
- 提供 Dockerfile 或 Docker Compose 文件
- 没有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
github/gh-aw 的其他 Issue
-
automated-analysis code-quality cookie testing
难度 2/5 1-3 小时 新手友好度 74/100
维护者通常 1 天内回复
-
wgdonenotdeferred: parenthesized Done() statement escapes non-deferred detection (paren_unwrap_gap)未关闭cookie sergo
难度 2/5 1-3 小时 新手友好度 76/100
维护者通常 1 天内回复
-
cookie eslint
难度 2/5 1-3 小时 新手友好度 75/100
维护者通常 1 天内回复
-
agentic-workflows
难度 5/5 一周以上 新手友好度 24/100
维护者通常 1 天内回复
-
agentic-workflows
难度 4/5 3-5 天 新手友好度 38/100
维护者通常 1 天内回复
相似的 Issue
-
area: records fast-track-debt priority: normal
难度 2/5 1-3 小时 新手友好度 76/100
维护者通常 1 天内回复
-
processor/genainormalizer
难度 2/5 1-3 小时 新手友好度 64/100
open-telemetry/opentelemetry-collector-contrib#52328 · 1 条评论 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 82/100
limitlessprojects/Go-Hybrid#1595 ·
维护者通常 1 天内回复
-
provider: hyper
难度 2/5 1-3 小时 新手友好度 78/100
charmbracelet/crush#4134 ·
维护者通常 1 天内回复
-
bug
难度 2/5 1-3 小时 新手友好度 66/100
AvengeMedia/DankMaterialShell#3912 ·