# ZeroBench leaderboard

Multimodal & Grounded 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/multimodal-grounded-zero-bench

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

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Multimodal & Grounded |
| Measure | Multi-step visual reasoning |
| Tasks | 100 visual reasoning questions |
| Difficulty | Tool-augmented visual reasoning |
| Published by | [Muse Spark Eval Methodology](https://ai.meta.com/static-resource/muse-spark-eval-methodology) |

A multi-step visual reasoning benchmark with pass@5 reporting and optional tool use.

## ZeroBench ranking

11 models with a published ZeroBench 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 | Multi-step visual reasoning | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Muse Spark Open](https://modelchoice.dev/models/muse-spark-open) | Meta | 50.0 | Observed 2026-11-21 · source benchlm:models |
| 2 | [GPT-5.4](https://modelchoice.dev/models/gpt-5-4) | OpenAI | 41.0 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Kimi K3.1 Pro](https://modelchoice.dev/models/kimi-3-1-pro) | Moonshot AI | 39.0 | Observed 2026-11-21 · source benchlm:models |
| 4 | [Muse Spark](https://modelchoice.dev/models/muse-spark) | Meta | 33.0 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Gemini 3.1 Pro](https://modelchoice.dev/models/gemini-3-1-pro) | Google | 29.0 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Qwen3.8 Max](https://modelchoice.dev/models/qwen3-8-max) | Alibaba | 24.0 | Observed 2026-11-21 · source benchlm:models |
| 7 | [Kimi K3](https://modelchoice.dev/models/kimi-3) | Moonshot AI | 23.0 | Observed 2026-11-21 · source benchlm:models |
| 8 | [dots3-note Preview](https://modelchoice.dev/models/dots3-note-preview) | Dots Studio | 19.0 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Seed 2.1 Pro](https://modelchoice.dev/models/seed-2-1-pro) | ByteDance | 18.0 | Observed 2026-11-21 · source benchlm:models |
| 10 | [Inkling 2 Mini](https://modelchoice.dev/models/inkling-2-mini) | Thinking Machines Lab | 17.0 | Observed 2026-11-21 · source benchlm:models |
| 11 | [Seed 2.1 Turbo](https://modelchoice.dev/models/seed-2-1-turbo) | ByteDance | 11.0 | 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.
