[dependabot]

Bumps @blockrun/llm from 3.23.1 to 3.24.4.

Release notes

Sourced from @​blockrun/llm's releases.

v3.24.4

Changed — router-core 8773a99

• @blockrun/router-core moves to BlockRunAI/router-core@8773a99. A live probe confirmed tool calls from deepseek/deepseek-v4.1-flash and openai/gpt-6.1-sol, so both join the verified tool fallbacks behind the configured chains. No primary changes.

Docs

• README model list adds google/gemini-4-argon (Gemini 4 Argon, live on the gateway on Base and Solana since 2026-10-20). Catalog only: routing tiers are unchanged until router-core lists it.

v3.24.3

Changed — router-core 2cd5a10

• @blockrun/router-core moves to BlockRunAI/router-core@2cd5a10, which adds capability rows for deepseek/deepseek-v4.1-flash, openai/gpt-6.1-sol and mistral/mistral-large-4. The routing algorithm is unchanged since 0.3.0.

v3.24.2

Changed — router-core 0.3.0

• @blockrun/router-core moves to 0.3.0 (BlockRunAI/router-core@65a92c8). Auto routing now reads only the user turn for structured-output, stakes, latency and language cues; a system prompt that lists tool schemas or says "be quick" no longer lifts a request off the cheapest tier. CJK input is sized at about one token per character, so a long Chinese document reaches the long-context models instead of overflowing a smaller window. Keyword scoring matches whole words ("classic" is no longer code, "history" is no longer creative), a stale speed probe fades toward neutral instead of toward zero, and a model with no catalog price no longer reports 100% savings.

v3.24.1

Changed — router-core 0.2.0

• @blockrun/router-core moves to 0.2.0 (BlockRunAI/router-core@27125a3). Eleven newly tool-verified models (Claude Fable 5.1 / Opus 5.5 / Sonnet 5.5, Gemini 3.8 Flash, GPT-5.1, GPT-6 Astra / Luna / Sol, Grok 4.6 / 4.7, MiMo 2.5) become availability fallbacks for tool-required turns. Primary selection is unchanged; on eco and auto the new fallbacks are tried cheapest first.

v3.24.0

Changed — smartChat with search falls back to an xAI model instead of failing

... (truncated)

Changelog

Sourced from @​blockrun/llm's changelog.

[3.24.4] - 2026-10-20

Changed — router-core 8773a99

• @blockrun/router-core moves to BlockRunAI/router-core@8773a99. A live probe confirmed tool calls from deepseek/deepseek-v4.1-flash and openai/gpt-6.1-sol, so both join the verified tool fallbacks behind the configured chains. No primary changes.

Docs

• README model list adds google/gemini-4-argon (Gemini 4 Argon, live on the gateway on Base and Solana since 2026-10-20). Catalog only: routing tiers are unchanged until router-core lists it.

[3.24.3] - 2026-10-20

Changed — router-core 2cd5a10

• @blockrun/router-core moves to BlockRunAI/router-core@2cd5a10, which adds capability rows for deepseek/deepseek-v4.1-flash, openai/gpt-6.1-sol and mistral/mistral-large-4. The routing algorithm is unchanged since 0.3.0.

[3.24.2] - 2026-10-18

Changed — router-core 0.3.0

• @blockrun/router-core moves to 0.3.0 (BlockRunAI/router-core@65a92c8). Auto routing now reads only the user turn for structured-output, stakes, latency and language cues; a system prompt that lists tool schemas or says "be quick" no longer lifts a request off the cheapest tier. CJK input is sized at about one token per character, so a long Chinese document reaches the long-context models instead of overflowing a smaller window. Keyword scoring matches whole words ("classic" is no longer code, "history" is no longer creative), a stale speed probe fades toward neutral instead of toward zero, and a model with no catalog price no longer reports 100% savings.

[3.24.1] - 2026-10-16

Changed — router-core 0.2.0

• @blockrun/router-core moves to 0.2.0 (BlockRunAI/router-core@27125a3). Eleven newly tool-verified models (Claude Fable 5.1 / Opus 5.5 / Sonnet 5.5, Gemini 3.8 Flash, GPT-5.1, GPT-6 Astra / Luna / Sol, Grok 4.6 / 4.7, MiMo 2.5) become availability fallbacks for tool-required turns. Primary selection is unchanged; on eco and auto the new fallbacks are tried cheapest first.

[3.24.0] - 2026-10-12

... (truncated)

Commits

• ceb43e8 release: 3.24.4 — router-core 8773a99 (#82)
• f6a1a6d docs(readme): list google/gemini-4-argon (#81)
• 2ebd748 release: 3.24.3 (#80)
• 1d76de4 Merge pull request #79 from BlockRunAI/chore/router-core-2cd5a10
• ad55fc0 chore: bump router-core to 2cd5a10
• ed40b1c chore: bump router-core to 0.3.0 (65a92c8) (#78)
• 9b92a55 release: 3.24.1
• c5ba92c chore: bump router-core to 0.2.0 (27125a3) (#77)
• 0abec11 Merge pull request #76 from BlockRunAI/feat/search-routing-xai-fallback
• 5816d9a Merge remote-tracking branch 'origin/test/typecheck-all-tests' into feat/sear...
• Additional commits viewable in compare view

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

Run configuration

• Source mode: Auto
• Repository: BlockRunAI/ClawRouter
• Branch: main
• File: .coderabbit.yaml
• Commit: 2e9acaf
• Layers: Repository YAML
• Review profile: CHILL
• Plan: Advanced
• Run ID: 5733bb90-69cf-4395-9ce5-0e61d8ae7700

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}