六家里三家的默认模型已经调不通了 —— 以及 README 那张表从此跟代码绑着 - #2
Merged
Merged
Conversation
2026-09 核对各家默认模型(官方文档经搜索核对,本机没有 key,没有真调用): kimi moonshot-v1-8k → kimi-k2.6 v1 整个系列 8/31 下线;K2.6 是 256K、面向 agent deepseek deepseek-chat → deepseek-flash 旧名 7/24 停用;deepseek-flash 是现在推荐的 V4.1 Flash gemini gemini-2.0-flash → gemini-3.5-flash 6/1 关停;agent.py 的 --model 注释里用的就是它 claude / openai / glm 三家的默认仍在服务,没动。 kimi-k2.6 默认开思考。Talos 回放历史时不带 reasoning_content —— Kimi 文档说 工具调用里不回传不会报错,只是推理连贯性可能差一点;Talos 也不设 tool_choice, 不撞「开思考时 tool_choice 只能 auto/none」那条。 新判据 test_the_provider_table_in_the_readme_is_what_the_code_does: README 的 provider 表(哪几家、读哪个 key、默认哪个模型)== agent.PROVIDERS。 这次两边一起改的;它管的是下一次只改一边。先只改代码,它当场点名三行,再改 README 才绿。 顺带:tests/test_memory.py 一段 docstring 里的 `\w` `\.` 是要原样显示的正则, 加 r 前缀 —— 3.12+ 的 SyntaxWarning / 3.11 的 DeprecationWarning 没了,值不变。 DEVELOPMENT.md 里 judge_live 的示例命令也从 deepseek-chat 换成 deepseek-flash。 337 判据:335 过、5 skip、1 xfail(原有),警告 0 条。 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014gdkHx6rLSMVsUmQiErVSn
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
默认模型
2026-09 核对各家默认模型(官方文档经搜索核对;本机没有 key,没有真调用):
moonshot-v1-8kkimi-k2.6deepseek-chatdeepseek-flashdeepseek-flash是现在推荐的 V4.1 Flashgemini-2.0-flashgemini-3.5-flashagent.py里--model的注释用的就是它claude / openai / glm 三家的默认仍在服务,没动。
kimi-k2.6默认开思考。Talos 回放历史时不带reasoning_content—— Kimi 文档说工具调用里不回传不会报错,只是推理连贯性可能差一点;Talos 也不设tool_choice,不撞「开思考时tool_choice只能 auto/none」那条。新判据
test_the_provider_table_in_the_readme_is_what_the_code_does:README 的 provider 表(哪几家、读哪个 key、默认哪个模型)==agent.PROVIDERS。先只改代码,它当场点名三行;再改 README 才绿。顺带
tests/test_memory.py一段 docstring 里的\w\.是要原样显示的正则,加r前缀:3.12+ 的 SyntaxWarning / 3.11 的 DeprecationWarning 没了,值不变DEVELOPMENT.md里 judge_live 的示例命令从deepseek-chat换成deepseek-flash验证
agent.py --selfcheck绿python agent.py --model kimi/kimi-k2.6 -p "列出当前目录"这种一句话验一下🤖 Generated with Claude Code
https://claude.ai/code/session_014gdkHx6rLSMVsUmQiErVSn
Generated by Claude Code