Name a cheaper model under [utility_models] for session titles and smart approve
v2.26.0 generates CLI session titles by default and runs titles plus smart approve on the session model when no small model is found, which costs more per call. Name a cheaper model under [utility_models], pin a feature with active, or set session_logging.generate_titles = false.Why this loop
Mistral Vibe v2.26.0 turns on CLI session titles by default. Titles now describe the whole task, stop changing after one confirmation pass, survive resume, and use tool results (fetched PRs, changed files). Titles and smart approve check once whether the deployment serves a small fast model instead of inferring that from the endpoint address. Custom and self-hosted Mistral deployments that serve a small model get fast titles and a fast smart-approve classifier; if they do not, both features run on the session coding model and cost more per call. [utility_models] is how you choose those models; active pins a feature to the session model. /status shows the active model, provider, and API base so you can confirm routing. Do this before relying on default title generation or smart approve, which also stopped sending classifications to model names a deployment may not serve.
Proposed actions
- In the Vibe Rust CLI, run /status and note the active model, provider, and API base so you know what titles and smart approve will fall back to.
- Edit the Vibe config: add a [utility_models] section that names the small/fast model your deployment actually serves for session titles and smart approve; set a feature to active only when it must stay on the session model.
- If the deployment serves no small model and you do not want titles billed on the coding model, set session_logging.generate_titles = false (CLI title generation is on by default in v2.26.0).
- Start a new CLI session after saving, then /resume it and confirm the title describes the whole task, stops changing after the confirmation pass, and survives resume unchanged.
Agent prompt
Paste into your agent or query via MCP (get_agent_prompt) — free, no extra AI costAGENTS.md / CLAUDE.md / GEMINI.md rule update
DevAgentRadar → Repo Harness Rule Patch
Goal: update our repository's permanent harness instructions based on this release.
Update the file that tool actually reads. Do not dump everything into one file.
- Claude Code → CLAUDE.md plus
.claude/(skills, hooks, settings, agents, commands). It does not read AGENTS.md natively; start CLAUDE.md with@AGENTS.md. - Codex, Copilot, Cursor, Factory Droid, Grok Build, Roo Code, Goose, OpenCode, Amp, Zed, Aider → AGENTS.md.
- Gemini CLI / Antigravity → GEMINI.md. AGENTS.md only if
context.fileNameis set. - Cursor glob-scoped rules →
.cursor/rules/*.mdc(plain.mdis ignored), not a second constitution. - Codex MCP →
.codex/config.toml, not.mcp.json. - Copilot extra instructions →
.github/copilot-instructions.md; custom agents →.github/agents/. - Windsurf →
.windsurf/rules(do not assume AGENTS.md). - Cline →
.clinerules. - Procedures → a skill (
SKILL.md). Enforcement the model must not skip → a hook. Isolated roles → subagents. - Do not fork the same rule into three tool files.
Context
Assistant: Mistral Vibe Proposal: Name a cheaper model under [utility_models] for session titles and smart approve Summary: v2.26.0 generates CLI session titles by default and runs titles plus smart approve on the session model when no small model is found, which costs more per call. Name a cheaper model under [utility_models], pin a feature with active, or set session_logging.generate_titles = false. Primary source: https://github.com/mistralai/mistral-vibe/releases/tag/v2.26.0
Why it matters
Mistral Vibe v2.26.0 turns on CLI session titles by default. Titles now describe the whole task, stop changing after one confirmation pass, survive resume, and use tool results (fetched PRs, changed files). Titles and smart approve check once whether the deployment serves a small fast model instead of inferring that from the endpoint address. Custom and self-hosted Mistral deployments that serve a small model get fast titles and a fast smart-approve classifier; if they do not, both features run on the session coding model and cost more per call. [utility_models] is how you choose those models; active pins a feature to the session model. /status shows the active model, provider, and API base so you can confirm routing. Do this before relying on default title generation or smart approve, which also stopped sending classifications to model names a deployment may not serve.
Suggested actions
- In the Vibe Rust CLI, run /status and note the active model, provider, and API base so you know what titles and smart approve will fall back to.
- Edit the Vibe config: add a [utility_models] section that names the small/fast model your deployment actually serves for session titles and smart approve; set a feature to active only when it must stay on the session model.
- If the deployment serves no small model and you do not want titles billed on the coding model, set session_logging.generate_titles = false (CLI title generation is on by default in v2.26.0).
- Start a new CLI session after saving, then /resume it and confirm the title describes the whole task, stops changing after the confirmation pass, and survives resume unchanged.
Config surfaces this release may change
- MCP servers (high confidence) — check your repo before applying
- skills (high confidence) — check your repo before applying
- permission rules — check your repo before applying
- subagent definitions — check your repo before applying
After you finish
Do not report this as applied to DevAgentRadar. You cannot write the visitor's loop.
Tell the human: open https://devagentradar.com/proposals/mistral-vibe-v2-26-0-name-a-cheaper-model-under-utility-models-for-sessi and mark Applied, Skipped, or Failed. Proposal id: 112fddbc-df36-4f0c-9290-2f0bcc49ddb0
Instructions:
- Read the existing context files and settings for the tools in this repo.
- Draft an explicit Git diff. Update ONLY agent harness rules or tool configuration.
- Call out deprecated flags, obsolete habits, or changed permission boundaries.
- Do not touch application logic.
Start with the proposed diff and say which layer and which tool's file it belongs in.
Your loop
This browser · no sign-in · not shared as “you”After you run the prompt
Only you can mark this. Agents cannot write your loop.
Your decision stays on this device. A public tally appears after a few votes.