From 21619ed44c16a0862606f9782f3901031c808f66 Mon Sep 17 00:00:00 2001 From: Bogdan Dolin Date: Sun, 19 Apr 2026 19:35:52 +0700 Subject: [PATCH 1/3] docs: update model references to Claude Opus 4.7 Align docs with the models actually wired up in the agents: - Opus 4.6 -> Opus 4.7 across modes, what-models, faq, how-does-it-work, and claude-code-comparison pages - Correct Thinker/Editor/Reviewer model lists in how-does-it-work.mdx to match the current agent definitions - Replace stale GPT-5.1 / Grok 4.1 / Relace mentions in faq.mdx with the Opus 4.7, GPT-5.4, and Gemini 3.1 Flash Lite setup currently in use Made-with: Cursor --- web/src/content/advanced/claude-code-comparison.mdx | 2 +- web/src/content/advanced/how-does-it-work.mdx | 8 ++++---- web/src/content/advanced/what-models.mdx | 12 ++++++------ web/src/content/help/faq.mdx | 2 +- web/src/content/tips/modes.mdx | 10 +++++----- 5 files changed, 17 insertions(+), 17 deletions(-) diff --git a/web/src/content/advanced/claude-code-comparison.mdx b/web/src/content/advanced/claude-code-comparison.mdx index 6048c6617b..b22dadf877 100644 --- a/web/src/content/advanced/claude-code-comparison.mdx +++ b/web/src/content/advanced/claude-code-comparison.mdx @@ -14,7 +14,7 @@ Both tools: - Run in your terminal - Understand your entire codebase context - Can edit files and execute terminal commands -- Use Claude models (Codebuff uses Claude Opus 4.6 as its orchestrator) +- Use Claude models (Codebuff uses Claude Opus 4.7 as its orchestrator) ## When to Choose Codebuff diff --git a/web/src/content/advanced/how-does-it-work.mdx b/web/src/content/advanced/how-does-it-work.mdx index 76e789def8..da5fd45e62 100644 --- a/web/src/content/advanced/how-does-it-work.mdx +++ b/web/src/content/advanced/how-does-it-work.mdx @@ -11,7 +11,7 @@ Codebuff runs multiple agents, each tuned for a specific task. ## The Orchestrator -The main agent ("Buffy") runs on Claude Opus 4.6. It reads your prompt, gathers context, and spawns subagents. The orchestrator is available in several variants: +The main agent ("Buffy") runs on Claude Opus 4.7. It reads your prompt, gathers context, and spawns subagents. The orchestrator is available in several variants: - [`base2`](/publishers/codebuff/agents/base2) - Default mode orchestrator - [`base2-lite`](/publishers/codebuff/agents/base2-lite) - Lite mode (faster, cheaper) @@ -23,9 +23,9 @@ The main agent ("Buffy") runs on Claude Opus 4.6. It reads your prompt, gathers - [**File Picker**](/publishers/codebuff/agents/file-picker) (Gemini 2.0 Flash) - finds relevant files - [**Code Searcher**](/publishers/codebuff/agents/code-searcher) - grep-style pattern matching - [**Researcher**](/publishers/codebuff/agents/researcher) (Grok 4 Fast) - web and docs lookup -- [**Thinker**](/publishers/codebuff/agents/thinker) (GPT-5.1, Gemini 2.5 Pro) - works through hard problems -- [**Editor**](/publishers/codebuff/agents/editor) (GPT-5.1, Claude Opus 4.6) - writes and modifies code -- [**Reviewer**](/publishers/codebuff/agents/reviewer) (Claude Opus 4.6, GLM 5.1 in Lite mode) - catches bugs and style issues +- [**Thinker**](/publishers/codebuff/agents/thinker) (Claude Opus 4.7, GPT-5.4) - works through hard problems +- [**Editor**](/publishers/codebuff/agents/editor) (Claude Opus 4.7, GPT-5.1, GLM 5.1) - writes and modifies code +- [**Reviewer**](/publishers/codebuff/agents/reviewer) (Claude Opus 4.7, GLM 5.1 in Lite mode) - catches bugs and style issues - [**Basher**](/publishers/codebuff/agents/basher) (Gemini 3.1 Flash Lite) - runs terminal commands ## Best-of-N Selection (Max Mode) diff --git a/web/src/content/advanced/what-models.mdx b/web/src/content/advanced/what-models.mdx index 19846213fa..5714254c52 100644 --- a/web/src/content/advanced/what-models.mdx +++ b/web/src/content/advanced/what-models.mdx @@ -16,9 +16,9 @@ The main agent ("Buffy") coordinates everything: | Mode | Model | |------|-------| - | Default | Opus 4.6 | - | Plan | Opus 4.6 | - | Max | Opus 4.6 | + | Default | Opus 4.7 | + | Plan | Opus 4.7 | + | Max | Opus 4.7 | | Free | GLM 5.1 | @@ -29,9 +29,9 @@ The orchestrator spawns these for specific jobs: | Task | Models | |------|--------| - | Code editing | Claude Opus 4.6, GLM 5.1 | - | Thinking/reasoning | Claude Opus 4.6, GPT-5.4 | - | Code review | Claude Opus 4.6, GPT-5.4 | + | Code editing | Claude Opus 4.7, GLM 5.1 | + | Thinking/reasoning | Claude Opus 4.7, GPT-5.4 | + | Code review | Claude Opus 4.7, GPT-5.4 | | File discovery | Gemini 3.1 Flash Lite, Gemini 2.5 Flash Lite | | Terminal commands | Gemini 3.1 Flash Lite | | Web/docs research | Gemini 3.1 Flash Lite | diff --git a/web/src/content/help/faq.mdx b/web/src/content/help/faq.mdx index 104ae35bc1..39940f640c 100644 --- a/web/src/content/help/faq.mdx +++ b/web/src/content/help/faq.mdx @@ -13,7 +13,7 @@ Software development: Writing features, tests, and scripts across common languag ## What model does Codebuff use? -Multiple. The orchestrator uses Claude Opus 4.6 in Default and Max modes, or GLM 5.1 in Free mode. Subagents are matched to their tasks: Claude Opus 4.6 for code editing, GPT-5.1 for deep reasoning, Grok 4.1 Fast for terminal commands and research, and Relace AI for fast file rewrites. Free mode includes code review support. See [What models do you use?](/docs/advanced/what-models) for the full breakdown. +Multiple. The orchestrator uses Claude Opus 4.7 in Default and Max modes, or GLM 5.1 in Free mode. Subagents are matched to their tasks: Claude Opus 4.7 (with GLM 5.1 as a lighter option) for code editing, Claude Opus 4.7 and GPT-5.4 for deep reasoning and code review, and Gemini 3.1 Flash Lite for terminal commands, file discovery, and web/docs research. Free mode includes code review support. See [What models do you use?](/docs/advanced/what-models) for the full breakdown. ## Can I use my Claude Pro or Max subscription with Codebuff? diff --git a/web/src/content/tips/modes.mdx b/web/src/content/tips/modes.mdx index b54d94e5a7..1b67daecd6 100644 --- a/web/src/content/tips/modes.mdx +++ b/web/src/content/tips/modes.mdx @@ -12,15 +12,15 @@ Codebuff has four modes. Switch during a session with `Shift+Tab` or `/mode:` co | Mode | Model | Editor Agent | Code Review | | --- | --- | --- | --- | --- | - | Default | Claude Opus 4.6 | editor | Yes | - | Max | Claude Opus 4.6 | editor-multi-prompt | Yes | - | Plan | Claude Opus 4.6 | None | No | + | Default | Claude Opus 4.7 | editor | Yes | + | Max | Claude Opus 4.7 | editor-multi-prompt | Yes | + | Plan | Claude Opus 4.7 | None | No | | Lite | GLM 5.1 | None | No | ## Default -Standard mode with Claude Opus 4.6: +Standard mode with Claude Opus 4.7: - Spawns [file pickers](/publishers/codebuff/agents/file-picker) and [code searchers](/publishers/codebuff/agents/code-searcher) to gather context - Uses the [`editor`](/publishers/codebuff/agents/editor) agent for code changes @@ -32,7 +32,7 @@ Switch to this mode with `/mode:default`. ## Max -Claude Opus 4.6 with best-of-N selection: +Claude Opus 4.7 with best-of-N selection: - Reads 12-20+ files per task - Spawns multiple [editor](/publishers/codebuff/agents/editor) agents in parallel, each with a different strategy From 27003bca09592712b17bc95b77dc557e9113faa2 Mon Sep 17 00:00:00 2001 From: Bogdan Dolin Date: Sun, 19 Apr 2026 19:46:49 +0700 Subject: [PATCH 2/3] docs: tighten FAQ model answer and fix SEO metadata - Rewrite the "What model does Codebuff use?" answer to only list modes that actually ship (Default, Max, Lite) and the models actually in use (Opus 4.7, GPT-5.4, Gemini 3.1 Flash Lite), dropping stale mentions of Free mode and Relace for file rewrites (file writes are handled deterministically by the runtime). - Update the mirrored FAQ entry in the docs page SEO structured data so it matches faq.mdx. Made-with: Cursor --- web/src/app/docs/[category]/[slug]/page.tsx | 2 +- web/src/content/help/faq.mdx | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/web/src/app/docs/[category]/[slug]/page.tsx b/web/src/app/docs/[category]/[slug]/page.tsx index b503269a62..cb442c7121 100644 --- a/web/src/app/docs/[category]/[slug]/page.tsx +++ b/web/src/app/docs/[category]/[slug]/page.tsx @@ -33,7 +33,7 @@ const FAQ_ITEMS = [ { question: 'What model does Codebuff use?', answer: - 'Multiple. The orchestrator uses Claude Opus 4.6 in Default and Max modes, or GLM 5.1 in Free mode. Subagents are matched to their tasks: Claude Opus 4.6 for code editing, GPT-5.1 for deep reasoning, Grok 4.1 Fast for terminal commands and research, and Relace AI for fast file rewrites.', + 'Multiple. The orchestrator uses Claude Opus 4.7 in Default and Max modes, or GLM 5.1 in Lite mode. Subagents are matched to their tasks: Claude Opus 4.7 and GPT-5.4 for deep reasoning and code review, and Gemini 3.1 Flash Lite for terminal commands, file discovery, and web/docs research.', }, { question: 'Can I use my Claude Pro or Max subscription with Codebuff?', diff --git a/web/src/content/help/faq.mdx b/web/src/content/help/faq.mdx index 39940f640c..477adbd8f5 100644 --- a/web/src/content/help/faq.mdx +++ b/web/src/content/help/faq.mdx @@ -13,7 +13,7 @@ Software development: Writing features, tests, and scripts across common languag ## What model does Codebuff use? -Multiple. The orchestrator uses Claude Opus 4.7 in Default and Max modes, or GLM 5.1 in Free mode. Subagents are matched to their tasks: Claude Opus 4.7 (with GLM 5.1 as a lighter option) for code editing, Claude Opus 4.7 and GPT-5.4 for deep reasoning and code review, and Gemini 3.1 Flash Lite for terminal commands, file discovery, and web/docs research. Free mode includes code review support. See [What models do you use?](/docs/advanced/what-models) for the full breakdown. +Multiple. The orchestrator uses Claude Opus 4.7 in Default and Max modes, or GLM 5.1 in Lite mode. Subagents are matched to their tasks: Claude Opus 4.7 and GPT-5.4 for deep reasoning and code review, and Gemini 3.1 Flash Lite for terminal commands, file discovery, and web/docs research. See [What models do you use?](/docs/advanced/what-models) for the full breakdown. ## Can I use my Claude Pro or Max subscription with Codebuff? From c157a1c13d86ec3f42a3e13eb19c4e1b12d2fb53 Mon Sep 17 00:00:00 2001 From: Bogdan Dolin Date: Sun, 19 Apr 2026 20:03:31 +0700 Subject: [PATCH 3/3] docs: fix review findings on model/mode consistency - what-models.mdx: rename "Free" -> "Lite" in the orchestrator table and the trailing paragraph so every doc in this PR uses the same user-facing mode label. - how-does-it-work.mdx: the Researcher subagent is Gemini 3.1 Flash Lite (agents/researcher/researcher-web.ts and researcher-docs.ts), not Grok 4 Fast. Matches what-models.mdx and faq.mdx. - page.tsx: update the FAQ JSON-LD answer for the Claude Pro / Max subscription question to reflect the deprecation that faq.mdx already surfaces, so SEO structured data and the rendered page agree. Made-with: Cursor --- web/src/app/docs/[category]/[slug]/page.tsx | 2 +- web/src/content/advanced/how-does-it-work.mdx | 2 +- web/src/content/advanced/what-models.mdx | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/web/src/app/docs/[category]/[slug]/page.tsx b/web/src/app/docs/[category]/[slug]/page.tsx index cb442c7121..44d5174e0a 100644 --- a/web/src/app/docs/[category]/[slug]/page.tsx +++ b/web/src/app/docs/[category]/[slug]/page.tsx @@ -38,7 +38,7 @@ const FAQ_ITEMS = [ { question: 'Can I use my Claude Pro or Max subscription with Codebuff?', answer: - "Yes! If you have a Claude Pro or Max subscription, you can connect it to Codebuff and use your subscription for Claude model requests. This lets you save credits while still benefiting from Codebuff's intelligent orchestration. Run /connect:claude in the CLI to link your subscription. Note: Using your Claude Pro/Max subscription in Codebuff is not officially supported by Anthropic.", + 'Connecting your Claude Pro or Max subscription to Codebuff is deprecated and will be removed on March 1st. At least one user had their Anthropic account disabled after heavy usage via Codebuff. We recommend switching to a Codebuff Strong subscription instead — it includes generous usage limits across all models without needing to connect an external subscription.', }, { question: 'Is Codebuff open source?', diff --git a/web/src/content/advanced/how-does-it-work.mdx b/web/src/content/advanced/how-does-it-work.mdx index da5fd45e62..08f13366f5 100644 --- a/web/src/content/advanced/how-does-it-work.mdx +++ b/web/src/content/advanced/how-does-it-work.mdx @@ -22,7 +22,7 @@ The main agent ("Buffy") runs on Claude Opus 4.7. It reads your prompt, gathers - [**File Picker**](/publishers/codebuff/agents/file-picker) (Gemini 2.0 Flash) - finds relevant files - [**Code Searcher**](/publishers/codebuff/agents/code-searcher) - grep-style pattern matching -- [**Researcher**](/publishers/codebuff/agents/researcher) (Grok 4 Fast) - web and docs lookup +- [**Researcher**](/publishers/codebuff/agents/researcher) (Gemini 3.1 Flash Lite) - web and docs lookup - [**Thinker**](/publishers/codebuff/agents/thinker) (Claude Opus 4.7, GPT-5.4) - works through hard problems - [**Editor**](/publishers/codebuff/agents/editor) (Claude Opus 4.7, GPT-5.1, GLM 5.1) - writes and modifies code - [**Reviewer**](/publishers/codebuff/agents/reviewer) (Claude Opus 4.7, GLM 5.1 in Lite mode) - catches bugs and style issues diff --git a/web/src/content/advanced/what-models.mdx b/web/src/content/advanced/what-models.mdx index 5714254c52..6fb3cd7367 100644 --- a/web/src/content/advanced/what-models.mdx +++ b/web/src/content/advanced/what-models.mdx @@ -19,7 +19,7 @@ The main agent ("Buffy") coordinates everything: | Default | Opus 4.7 | | Plan | Opus 4.7 | | Max | Opus 4.7 | - | Free | GLM 5.1 | + | Lite | GLM 5.1 | ## Subagents @@ -37,4 +37,4 @@ The orchestrator spawns these for specific jobs: | Web/docs research | Gemini 3.1 Flash Lite | -Max mode runs multiple implementations in parallel and picks the best one. Default mode runs a single implementation pass. Free mode uses GLM 5.1 and includes code review support. +Max mode runs multiple implementations in parallel and picks the best one. Default mode runs a single implementation pass. Lite mode uses GLM 5.1 and includes code review support.