补齐 Astra 元数据与 Max/Ultra 支持 / Add Astra metadata and reasoning efforts - #2112
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
中文
问题与范围
gpt-6-astra尚未进入内置兼容元数据查询,使用自定义供应商时会回落到通用模型信息,缺少原生max/ultra推理档位。用户已通过本地 catalog 配置验证这两个档位可用;此 PR 将对应元数据单独补齐。low / medium / high / xhigh / max / ultra,默认medium,默认上下文窗口 272K。additional_speed_tiers: ["fast"],沿用现有priorityservice tier,使 catalog 与 UI 元数据一致。gpt-6-astra,不猜测别名。推理能力字段依据本地工作的兼容配置;272K 为本补丁的默认值。未将这些兼容值宣称为独立核验的官方模型规格。Responses Lite 保持关闭。
Fast 能力依据 OpenAI Fast mode 文档:请求接受
service_tier: "fast"或兼容值"priority"。本补丁沿用 Codex++ 现有priority通路,不修改请求逻辑。官方注明 Astra 的 Fast 不包含延迟 SLA,且不支持 EU 数据驻留;第三方供应商仍需自行支持此能力。验证
cargo test --offline -q -p codex-plus-core --test model_suffix --test relay_config -j 2 -- --test-threads=1model_suffix: 16 passedrelay_config: 146 passedgit diff --check: passed覆盖 catalog/UI 六档及 Fast 能力一致性、272K 默认值、显式窗口覆盖、未知别名隔离及 PureApi 自动生成 catalog。未安装此分支产物,也未重新验证真实 API 请求;真人验证来自已有本地配置。
English
Add an exact-match Astra compatibility metadata entry through the existing catalog and UI metadata lookup. Expose native
maxandultraalongside low/medium/high/xhigh, with a medium default and a 272K default context window.This is a metadata-only fix: no new reasoning controls, request rewriting, proxy behavior or installation changes. Explicit window overrides and existing model behavior remain intact. Capability values follow a locally working catalog configuration, not independently verified official model specifications; the user's personal 300K override is deliberately excluded.
Fast support is separately verified against the official Fast mode guide, which accepts both
fastandpriority. Declareadditional_speed_tiers: ["fast"]and reuse the existingpriorityservice tier. Astra Fast has no latency SLA and is unavailable with EU data residency; third-party provider support is still required.Validation: 16 model-suffix tests and 146 relay-config tests pass, including catalog/UI effort and Fast parity and no-suffix PureApi catalog generation.
git diff --checkpasses. No build from this branch was installed and no fresh real API request was tested.