SlackBench - CLI - Overall: leaderboard
Agents answer questions across realistic Slack workplace reenactments and associated evolving code repositories. Mean category pass score (%). SlackBench140tasks:58AgenticSearch,48Hallucination,14Summarization,20CircumstanceInference; three runs,20min/query; GPT-5.4 judge. Summarization earns checklist partial credit, other categories binary pass; Overall is the unweighted category mean. Harness, MCP/CLI access and reasoning effort remain separate in row identities. Native first-party model APIs.
Metric: Mean category pass score (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 (Benchmark Sweep Run: source=SlackBench; harness=Claude Code; access=CLI; effort=medium) | 73.68 | |
| 2 | GPT-5.5 (Benchmark Sweep Run: source=SlackBench; harness=Codex CLI; access=CLI; effort=medium) | 59.9 | |
| 3 | Claude Opus 4.7 (Benchmark Sweep Run: source=SlackBench; harness=Claude Code; access=CLI; effort=medium) | 59.63 | |
| 4 | Claude Sonnet 4.6 (Benchmark Sweep Run: source=SlackBench; harness=Claude Code; access=CLI; effort=medium) | 54.78 | |
| 5 | Gemini 3.1 Pro (Preview) (Benchmark Sweep Run: source=SlackBench; harness=Gemini CLI; access=CLI; effort=medium) | 54.31 | |
| 6 | GPT-5.4 Mini (Benchmark Sweep Run: source=SlackBench; harness=Codex CLI; access=CLI; effort=medium) | 43.12 | |
| 7 | Gemini 3 Flash (Preview) (Benchmark Sweep Run: source=SlackBench; harness=Gemini CLI; access=CLI; effort=medium) | 34.55 |
Interactive version: theaggregate.ai/benchmark?slug=slackbench-cli-overall · How It Works · Data refreshed daily, snapshot 2026-11-06.