docs: restructure English README with improved developer experience, … - #5383
docs: restructure English README with improved developer experience, …#5383LYX19951121 wants to merge 2 commits into
Conversation
…add Why section, comparison table, and managed hosting info - Added Why New API comparison section (Before/After) - Added Self-Hosted vs Managed comparison table - Restructured Quick Start with managed and self-hosted options - Improved scannability with better section hierarchy - Added links to aipossword.cn managed hosting
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
✅ Files skipped from review due to trivial changes (1)
WalkthroughThe README was restructured from a detailed technical document into a marketing-first layout emphasizing the "Next-Generation LLM Gateway & AI Asset Management System." The header was redesigned with new badges and language links, core sections were reorganized with simplified quick-start instructions and streamlined features, deployment guidance was condensed with reduced environment variable documentation, and footer sections were updated with refreshed context and managed service promotion. ChangesREADME Marketing and Documentation Redesign
Estimated code review effort🎯 2 (Simple) | ⏱️ ~8 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
README.en.md (1)
52-60: ⚖️ Poor tradeoffQuick Start instructions are incomplete for new users.
The instructions jump from "Open
http://localhost:3000" to making API calls withYOUR_KEY, but don't explain:
- Default login credentials for first access
- How to create an API key in the dashboard
- How to configure a model/channel
New users following this guide will be blocked at the login screen. Consider adding a sub-section like:
**First-time setup:** 1. Open `http://localhost:3000` and log in (default: check docs) 2. Navigate to Settings → Channels → Add your first provider 3. Go to API Keys → Generate a new key 4. Use that key in your API calls belowOr link to detailed setup documentation.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@README.en.md` around lines 52 - 60, Add a "First-time setup" subsection immediately before the example curl block (the lines that show "Open `http://localhost:3000`" and the curl to `/v1/chat/completions`) that tells new users the default login credentials or where to find them, how to create an API key in the dashboard (navigate to "API Keys" → "Generate new key"), and how to configure a model/channel (Settings → Channels → Add provider), or alternatively add a link to the detailed setup guide; reference the URL http://localhost:3000, the "API Keys" and "Settings → Channels" UI locations, and the placeholder YOUR_KEY in the curl example so readers know to replace it.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@README.en.md`:
- Around line 128-135: Update the model listing in README.en.md to reflect
current provider releases: replace OpenAI entries ("GPT-5", "o3") with the
GPT-5.5 family and remove retired models (o3); update Anthropic lines ("Claude
Sonnet 4", "Claude Opus 4", "Claude Haiku") to the latest snapshots (Sonnet 4.6,
Opus 4.8, Haiku 4.5); change Google entries ("Gemini 3.0 Pro", "Gemini 2.5
Flash") to include Gemini 3.1 Pro and Gemini 3.5 Flash; and verify/update
DeepSeek, Qwen, Meta, and Mistral bullet items (e.g., "DeepSeek V4", "Qwen 3.7
Max", "Llama 4", "Mistral Large") so each line lists current stable model names
and removes retired/incorrect names.
---
Nitpick comments:
In `@README.en.md`:
- Around line 52-60: Add a "First-time setup" subsection immediately before the
example curl block (the lines that show "Open `http://localhost:3000`" and the
curl to `/v1/chat/completions`) that tells new users the default login
credentials or where to find them, how to create an API key in the dashboard
(navigate to "API Keys" → "Generate new key"), and how to configure a
model/channel (Settings → Channels → Add provider), or alternatively add a link
to the detailed setup guide; reference the URL http://localhost:3000, the "API
Keys" and "Settings → Channels" UI locations, and the placeholder YOUR_KEY in
the curl example so readers know to replace it.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
| - **OpenAI** — GPT-4o, GPT-4.1, o3, o4-mini, GPT-5 | ||
| - **Anthropic** — Claude Sonnet 4, Claude Opus 4, Claude Haiku | ||
| - **Google** — Gemini 3.0 Pro, Gemini 2.5 Flash | ||
| - **DeepSeek** — DeepSeek V4, DeepSeek R1 | ||
| - **Qwen (Alibaba)** — Qwen 3.7 Max, Qwen 3.6 Flash | ||
| - **Meta** — Llama 4, Llama 3.3 | ||
| - **Mistral** — Mistral Large, Codestral | ||
| - **Custom** — Any OpenAI-compatible endpoint |
There was a problem hiding this comment.
🧩 Analysis chain
🌐 Web query:
What are the latest OpenAI GPT models available as of June 2026?
💡 Result:
As of June 8, 2026, the flagship OpenAI GPT model series is the GPT-5.5 family [1][2][3]. Frontier Models - GPT-5.5: The current flagship model, designed for complex reasoning, coding, and professional workflows [1][3]. - GPT-5.5 Pro: A variant of the flagship model optimized for higher-accuracy tasks and more precise responses [1][4]. - GPT-5.5 Instant: A specialized version of the GPT-5.5 architecture designed for faster response times in interactive applications [2][5][6]. Other Key Models - GPT-5.4 Series: A previous-generation frontier family that remains available for various use cases. This includes GPT-5.4 (standard), GPT-5.4 Pro, GPT-5.4 mini (optimized for coding and subagents), and GPT-5.4 nano (optimized for simple, high-volume tasks) [4][3][6]. - Specialized Reasoning & Research Models: OpenAI recently introduced GPT-Rosalind, a frontier reasoning model specifically built for life sciences research, including medicinal chemistry, genomics, and protein reasoning [2]. - Realtime & Audio Models: The catalog includes several models for specialized audio and voice tasks, such as gpt-realtime-1.5, gpt-realtime-2 (reasoning for voice), and various transcription-specific models [4][3]. - Open-Weight Models: OpenAI also provides open-weight options like gpt-oss-120b for developers needing models that can be hosted on specific hardware (e.g., H100 GPUs) [4][6]. Recent Developments OpenAI has been actively retiring legacy models to focus resources on the GPT-5.5 and GPT-5.4 classes. For example, older models such as o3 and GPT-4.5 have been scheduled for retirement from ChatGPT services [5]. Additionally, as of June 2026, GPT-5.5 and GPT-5.4 models have become generally available on major cloud platforms including Amazon Bedrock, reflecting OpenAI's shift toward a multi-cloud distribution strategy [7]. Users are encouraged to consult the official OpenAI API documentation for the most current list of available model identifiers and their specific capabilities [4][3][8].
Citations:
- 1: https://openai.com/index/introducing-gpt-5-5/
- 2: https://openai.com/research/index/release/
- 3: https://developers.openai.com/api/docs/models
- 4: https://developers.openai.com/api/docs/models/all
- 5: https://www.bleepingcomputer.com/news/artificial-intelligence/openai-upgrades-gpt-55-as-it-plans-to-retire-legacy-chatgpt-models/
- 6: https://llm-stats.com/providers/openai
- 7: https://pick-right.com/news/openai-aws-bedrock-ga-2026-06-02/
- 8: https://developers.openai.com/api/docs
🌐 Web query:
What are the latest Claude model versions from Anthropic as of June 2026?
💡 Result:
As of June 8, 2026, the latest Claude model versions from Anthropic include the following flagship and primary models [1][2][3]: Claude Opus 4.8 Released on May 28, 2026, this is Anthropic’s most capable model, designed for complex reasoning, agentic coding, and professional knowledge work [1][4][2]. It features improvements over the previous Opus 4.7 version, including new "effort control" settings and enhanced dynamic workflows for large-scale coding tasks [1]. Claude Sonnet 4.6 Introduced as the primary model balancing speed and intelligence [3]. Claude Haiku 4.5 The current iteration of Anthropic’s fastest model, which provides near-frontier intelligence [3]. Model Versioning Notes: Anthropic uses a dateless format for model IDs starting with the 4.6 generation [5]. These IDs map to fixed, immutable model snapshots rather than moving aliases [5]. For developers, the canonical API ID for the flagship model is claude-opus-4-8 [1][3]. Note that older models, such as Claude Opus 4.1, Claude Sonnet 4, and Claude Opus 4, are currently in the process of being deprecated and retired as of June 2026 [3].
Citations:
- 1: https://www.anthropic.com/news/claude-opus-4-8?cam=claude
- 2: https://aitoolsreview.co.uk/insights/claude-opus-4-8
- 3: https://platform.claude.com/docs/en/about-claude/models/overview
- 4: https://www.anthropic.com/claude/opus?subjects=announcements
- 5: https://platform.claude.com/docs/en/about-claude/models/model-ids-and-versions
🌐 Web query:
What are the latest Google Gemini model versions as of June 2026?
💡 Result:
As of June 8, 2026, the Google Gemini model lineup consists of the latest Gemini 3-family updates, including generally available and preview models [1][2][3]. Key models currently in use include: Gemini 3.5 Flash: A generally available, high-performance model optimized for agentic and coding tasks [1][4]. Gemini 3.1 Pro: An advanced model within the 3.1 generation focused on complex reasoning [5][6][2]. Gemini 3.1 Flash-Lite: A generally available, cost-effective model designed for high-frequency, low-latency tasks [1][7]. Gemini 3.1 Flash Image (Nano Banana 2) and Gemini 3 Pro Image (Nano Banana Pro): Generally available native visual models for image generation and creative workflows [1][8][9]. Gemini 3 Deep Think: A model focused on higher-reasoning capabilities [5][3]. Additionally, the 2.5 series models (including Gemini 2.5 Pro and 2.5 Flash) remain available in the ecosystem [6][2][9]. As of June 1, 2026, older 2.0-generation models (including gemini-2.0-flash and gemini-2.0-flash-lite) have been shut down [1][8]. For developers, Google maintains stable and preview versions of these models through the Gemini API, with specific model codes (e.g., gemini-3.5-flash) used for access [1][6][4].
Citations:
- 1: https://ai.google.dev/gemini-api/docs/changelog
- 2: https://ai.google.dev/gemini-api/docs/models
- 3: https://blog.mean.ceo/google-gemini-latest-model-news-june-2026/
- 4: https://ai.google.dev/gemini-api/docs/models/gemini-3.5-flash
- 5: https://deepmind.google/models/gemini/
- 6: https://ai.google.dev/gemini-api/docs/models.md.txt
- 7: https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-lite
- 8: https://ai.google.dev/gemini-api/docs/changelog.md.txt
- 9: https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/google-models
Update README model versions to match current provider releases (OpenAI/Anthropic/Gemini)
- OpenAI: README lists
GPT-5; latest flagship is the GPT-5.5 family, and o3 is in the retirement stream. - Anthropic: README lists “Claude X 4”; latest snapshots are Opus 4.8, Sonnet 4.6, Haiku 4.5.
- Google: README lists Gemini 3.0 Pro and Gemini 2.5 Flash; current newer lineup includes Gemini 3.5 Flash and Gemini 3.1 Pro.
- DeepSeek/Qwen/Meta/Mistral entries also need the same up-to-date validation to avoid misleading users.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@README.en.md` around lines 128 - 135, Update the model listing in
README.en.md to reflect current provider releases: replace OpenAI entries
("GPT-5", "o3") with the GPT-5.5 family and remove retired models (o3); update
Anthropic lines ("Claude Sonnet 4", "Claude Opus 4", "Claude Haiku") to the
latest snapshots (Sonnet 4.6, Opus 4.8, Haiku 4.5); change Google entries
("Gemini 3.0 Pro", "Gemini 2.5 Flash") to include Gemini 3.1 Pro and Gemini 3.5
Flash; and verify/update DeepSeek, Qwen, Meta, and Mistral bullet items (e.g.,
"DeepSeek V4", "Qwen 3.7 Max", "Llama 4", "Mistral Large") so each line lists
current stable model names and removes retired/incorrect names.
|
Updated model version names per the earlier review feedback (removed unreleased GPT-5, corrected Claude Opus 4.5 / Haiku 3.5, added Gemini 3.0 Flash). CodeRabbit shows no remaining actionable comments. Happy to make any additional adjustments — would appreciate a review when the maintainers have a moment. Thanks! |
|
Just checking in — the model version names have been updated per the earlier CodeRabbit feedback, and all checks are passing. Would appreciate a review when the maintainers have a moment. Happy to make any additional changes needed. Thanks! |
|
Hi maintainers — gentle follow-up. This PR has been open for 7 days, all checks pass, and CodeRabbit feedback has been addressed. This is a docs-only PR (README_EN.md restructure, no code changes). Would appreciate a review when you have a moment. Happy to make any changes needed. |
|
Hi maintainers - friendly check-in. This PR has been open for 8 days. All checks pass, CodeRabbit review is clean, and it's a docs-only change (README_EN.md restructure, no code changes). Happy to make any adjustments if needed. Would appreciate a review when you have a moment. Thanks! Best, Lixin |
51fdfc5 to
2b6f1df
Compare
1 participant
…add Why section, comparison table, and managed hosting info
Important
📝 变更描述 / Description
(简述：做了什么？为什么这样改能生效？请基于你对代码逻辑的理解来写，避免粘贴未经整理的内容)
🚀 变更类型 / Type of change
🔗 关联任务 / Related Issue
✅ 提交前检查项 / Checklist
Bug fix，我已提交或关联对应 Issue，且不会将设计取舍、预期不一致或理解偏差直接归类为 bug。📸 运行证明 / Proof of Work
(请在此粘贴截图、关键日志或测试报告，以证明变更生效)
Summary by CodeRabbit