# ApprenticeBench 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://modelscale.dev/leaderboards/benchmarks/agentic-apprentice-bench

ApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable job. Every model the catalog carries a published ApprenticeBench value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Agentic |
| Measure | Cumulative success rate over 100 bills |
| Tasks | 100 vendor bills processed in sequence inside a simulated construction company |
| Difficulty | Long-horizon computer use with offline and online continual learning |
| Published by | [ApprenticeBench: a step change in AI's job readiness](https://neocognition.io/blog/apprentice-bench/) |

Tests whether a computer-use agent can learn a real accounts-payable job on the job, processing 100 vendor bills in a company ERP system with only the handbook, historical records, and mentor feedback a new hire would get.

## ApprenticeBench ranking

18 models with a published ApprenticeBench 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 | Cumulative success rate over 100 bills | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Claude Fable 5.1](https://modelscale.dev/models/claude-fable-5-1) | Anthropic | 72 | Observed 2026-11-21 · source benchlm:models |
| 2 | [GPT-6 Astra](https://modelscale.dev/models/gpt-6-astra) | OpenAI | 68 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Claude Opus 5](https://modelscale.dev/models/claude-opus-5) | Anthropic | 36 | Observed 2026-11-21 · source benchlm:models |
| 4 | [Claude Fable 5](https://modelscale.dev/models/claude-fable-5) | Anthropic | 34 | Observed 2026-11-21 · source benchlm:models |
| 5 | [GPT-5.6 Sol](https://modelscale.dev/models/gpt-5-6-sol) | OpenAI | 26 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Gemini 3.8 Flash](https://modelscale.dev/models/gemini-3-8-flash) | Google | 24 | Observed 2026-11-21 · source benchlm:models |
| 7 | [GPT-5.5](https://modelscale.dev/models/gpt-5-5) | OpenAI | 20 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Muse Spark 1.3](https://modelscale.dev/models/muse-spark-1-3) | Meta | 19 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Kimi K3](https://modelscale.dev/models/kimi-3) | Moonshot AI | 18 | Observed 2026-11-21 · source benchlm:models |
| 10 | [Claude Sonnet 5](https://modelscale.dev/models/claude-sonnet-5) | Anthropic | 16 | Observed 2026-11-21 · source benchlm:models |
| 10 | [Gemini 3.7 Flash](https://modelscale.dev/models/gemini-3-7-flash) | Google | 16 | Observed 2026-11-21 · source benchlm:models |
| 10 | [GPT-5.6 Terra](https://modelscale.dev/models/gpt-5-6-terra) | OpenAI | 16 | Observed 2026-11-21 · source benchlm:models |
| 13 | [Grok 4.6](https://modelscale.dev/models/grok-4-6) | xAI | 13 | Observed 2026-11-21 · source benchlm:models |
| 14 | [GPT-5.4](https://modelscale.dev/models/gpt-5-4) | OpenAI | 11 | Observed 2026-11-21 · source benchlm:models |
| 15 | [Claude Opus 4.7](https://modelscale.dev/models/claude-opus-4-7) | Anthropic | 7 | Observed 2026-11-21 · source benchlm:models |
| 15 | [GPT-5.6 Luna](https://modelscale.dev/models/gpt-5-6-luna) | OpenAI | 7 | Observed 2026-11-21 · source benchlm:models |
| 17 | [Claude Opus 4.6](https://modelscale.dev/models/claude-opus-4-6) | Anthropic | 5 | Observed 2026-11-21 · source benchlm:models |
| 18 | [Claude Sonnet 4.6](https://modelscale.dev/models/claude-sonnet-4-6) | Anthropic | 2 | Observed 2026-11-21 · 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.
