# JobBench leaderboard

Agentic benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/agentic-job-bench

Every model the catalog carries a published JobBench value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Agentic |
| Measure | Agentic workplace deliverables |
| Tasks | 130 tasks across 35 occupations |
| Difficulty | Professional multi-source workflows |
| Published by | [JobBench: Aligning Agent Work With Human Will](https://arxiv.org/abs/2605.26329) |

An occupational agent benchmark for professional workflows that workers say they most want delegated to AI.

## JobBench ranking

39 models with a published JobBench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Agentic workplace deliverables | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Muse Wonder Full](https://modelchoice.dev/models/muse-wonder-full) | Meta | 75.3 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Inkling 2 Small](https://modelchoice.dev/models/inkling-2-small) | Thinking Machines Lab | 70.5 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Muse Wonder](https://modelchoice.dev/models/muse-wonder) | Meta | 70.5 | Observed 2026-11-21 · source benchlm:models |
| 4 | [LongCat-2.5](https://modelchoice.dev/models/longcat-2-5) | Meituan | 65.4 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Muse Spark 1.3](https://modelchoice.dev/models/muse-spark-1-3) | Meta | 64.9 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Ultramarine 1 Flash](https://modelchoice.dev/models/ultramarine-1-flash) | Paper Instruments | 62.9 | Observed 2026-11-21 · source benchlm:models |
| 7 | [MiMo-V2.6-Pro](https://modelchoice.dev/models/mimo-v2-6-pro) | Xiaomi | 62.0 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Muse Spark Open](https://modelchoice.dev/models/muse-spark-open) | Meta | 61.8 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Hy4 preview](https://modelchoice.dev/models/hy4-preview) | Tencent | 61.7 | Observed 2026-11-21 · source benchlm:models |
| 10 | [MiMo-V2.6-Flash](https://modelchoice.dev/models/mimo-v2-6-flash) | Xiaomi | 61.2 | Observed 2026-11-21 · source benchlm:models |
| 11 | [Step 5 Preview](https://modelchoice.dev/models/step-5-preview) | StepFun | 59.0 | Observed 2026-11-21 · source benchlm:models |
| 12 | [Qwen3.8-Flash-Next](https://modelchoice.dev/models/qwen3-8-flash-next) | Alibaba | 55.7 | Observed 2026-11-21 · source benchlm:models |
| 13 | [Muse Spark 1.1](https://modelchoice.dev/models/muse-spark-1-1) | Meta | 54.7 | Observed 2026-11-21 · source benchlm:models |
| 14 | [Qwen3.8 Max](https://modelchoice.dev/models/qwen3-8-max) | Alibaba | 53.4 | Observed 2026-11-21 · source benchlm:models |
| 15 | [Kimi K3](https://modelchoice.dev/models/kimi-3) | Moonshot AI | 52.9 | Observed 2026-11-21 · source benchlm:models |
| 16 | [Atria Dawn Preview](https://modelchoice.dev/models/atria-dawn-preview) | Shanghai Artificial Intelligence Laboratory | 50.3 | Observed 2026-11-21 · source benchlm:models |
| 17 | [Inkling 2 Mini](https://modelchoice.dev/models/inkling-2-mini) | Thinking Machines Lab | 49.6 | Observed 2026-11-21 · source benchlm:models |
| 18 | [Claude Opus 4.7](https://modelchoice.dev/models/claude-opus-4-7) | Anthropic | 45.9 | Observed 2026-11-21 · source benchlm:models |
| 18 | [Claude Opus 4.7 (Adaptive)](https://modelchoice.dev/models/claude-opus-4-7-max) | Anthropic | 45.9 | Last good 2026-11-18 · source benchlm:models |
| 20 | [GPT-5.5](https://modelchoice.dev/models/gpt-5-5) | OpenAI | 42.7 | Observed 2026-11-21 · source benchlm:models |
| 21 | [GPT-5.4](https://modelchoice.dev/models/gpt-5-4) | OpenAI | 38.9 | Observed 2026-11-21 · source benchlm:models |
| 22 | [Claude Sonnet 4.6](https://modelchoice.dev/models/claude-sonnet-4-6) | Anthropic | 36.9 | Observed 2026-11-21 · source benchlm:models |
| 23 | [Claude Opus 4.6](https://modelchoice.dev/models/claude-opus-4-6) | Anthropic | 36.7 | Observed 2026-11-21 · source benchlm:models |
| 24 | [GPT-5.2](https://modelchoice.dev/models/gpt-5-2) | OpenAI | 34.3 | Observed 2026-11-21 · source benchlm:models |
| 25 | [GPT-5.3 Codex](https://modelchoice.dev/models/gpt-5-3-codex) | OpenAI | 33.7 | Observed 2026-11-21 · source benchlm:models |
| 26 | [Qwen3.8-27B](https://modelchoice.dev/models/qwen3-8-27b) | Alibaba | 33.4 | Observed 2026-11-21 · source benchlm:models |
| 27 | [Claude Opus 4.5](https://modelchoice.dev/models/claude-opus-4-5) | Anthropic | 32.3 | Observed 2026-11-21 · source benchlm:models |
| 28 | [Claude Sonnet 4.5](https://modelchoice.dev/models/claude-sonnet-4-5) | Anthropic | 27.7 | Observed 2026-11-21 · source benchlm:models |
| 29 | [GPT-5.1-Codex](https://modelchoice.dev/models/gpt-5-1-codex) | OpenAI | 26.2 | Observed 2026-11-21 · source benchlm:models |
| 30 | [GPT-5.2-Codex](https://modelchoice.dev/models/gpt-5-2-codex) | OpenAI | 26.0 | Observed 2026-11-21 · source benchlm:models |
| 31 | [Claude 4.1 Opus](https://modelchoice.dev/models/claude-4-1-opus) | Anthropic | 21.9 | Observed 2026-11-21 · source benchlm:models |
| 32 | [Qwen3.5 Plus](https://modelchoice.dev/models/qwen3-5-plus) | Alibaba | 18.5 | Observed 2026-11-21 · source benchlm:models |
| 33 | [Claude 4 Sonnet](https://modelchoice.dev/models/claude-4-sonnet) | Anthropic | 18.4 | Observed 2026-11-21 · source benchlm:models |
| 34 | [Claude Haiku 4.5](https://modelchoice.dev/models/claude-haiku-4-5) | Anthropic | 16.0 | Observed 2026-11-21 · source benchlm:models |
| 35 | [Gemini 3 Flash](https://modelchoice.dev/models/gemini-3-flash) | Google | 11.4 | Observed 2026-11-21 · source benchlm:models |
| 35 | [Gemini 3 Pro](https://modelchoice.dev/models/gemini-3-pro) | Google | 11.4 | Observed 2026-11-21 · source benchlm:models |
| 37 | [Kimi K2.5](https://modelchoice.dev/models/kimi-k2-5) | Moonshot AI | 8.7 | Observed 2026-11-21 · source benchlm:models |
| 38 | [GPT-5](https://modelchoice.dev/models/gpt-5) | OpenAI | 8.5 | Observed 2026-11-21 · source benchlm:models |
| 38 | [GPT-5 (high)](https://modelchoice.dev/models/gpt-5-high) | OpenAI | 8.5 | Last good 2026-11-18 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
