# VulcanBench Frontier v4 leaderboard

Coding 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/coding-vulcan-bench-v4

Every model the catalog carries a published VulcanBench Frontier v4 value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Coding |
| Measure | Combined score (correctness and judged code quality), best effort per model |
| Tasks | 23 behavioural-reconstruction coding tasks |
| Difficulty | Professional multi-file software engineering |
| Published by | [VulcanBench Frontier v4 leaderboard](https://vulcanbench.com/leaderboard.html) |

An open coding benchmark of 23 hard behavioural-reconstruction tasks, run at every reasoning-effort level and scored on correctness and judged code quality.

## VulcanBench Frontier v4 ranking

19 models with a published VulcanBench Frontier v4 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 | Combined score (correctness and judged code quality), best effort per model | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Claude Fable 5.2](https://modelchoice.dev/models/claude-fable-5-2) | Anthropic | 93.8 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Grok 4.7](https://modelchoice.dev/models/grok-4-7) | xAI | 93.2 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Claude Haiku 5.5](https://modelchoice.dev/models/claude-haiku-5-5) | Anthropic | 92.2 | Observed 2026-11-21 · source benchlm:models |
| 4 | [Claude Sonnet 5.5](https://modelchoice.dev/models/claude-sonnet-5-5) | Anthropic | 92.0 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Claude Fable 5.1](https://modelchoice.dev/models/claude-fable-5-1) | Anthropic | 91.8 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Claude Opus 5.5](https://modelchoice.dev/models/claude-opus-5-5) | Anthropic | 91.1 | Observed 2026-11-21 · source benchlm:models |
| 7 | [GPT-6.1 Terra](https://modelchoice.dev/models/gpt-6-1-terra) | OpenAI | 90.0 | Observed 2026-11-21 · source benchlm:models |
| 8 | [GPT-6 Astra](https://modelchoice.dev/models/gpt-6-astra) | OpenAI | 89.3 | Observed 2026-11-21 · source benchlm:models |
| 9 | [GPT-5.6 Terra](https://modelchoice.dev/models/gpt-5-6-terra) | OpenAI | 89.0 | Observed 2026-11-21 · source benchlm:models |
| 10 | [GPT-6.2 Sol](https://modelchoice.dev/models/gpt-6-2-sol) | OpenAI | 89.0 | Observed 2026-11-21 · source benchlm:models |
| 11 | [GPT-6.1 Sol](https://modelchoice.dev/models/gpt-6-1-sol) | OpenAI | 88.8 | Observed 2026-11-21 · source benchlm:models |
| 12 | [DeepSeek V4.1 Flash 1017](https://modelchoice.dev/models/deepseek-v4-1-flash) | DeepSeek | 88.7 | Observed 2026-11-21 · source benchlm:models |
| 13 | [Muse Wonder Full](https://modelchoice.dev/models/muse-wonder-full) | Meta | 88.4 | Observed 2026-11-21 · source benchlm:models |
| 14 | [GPT-5.6 Sol](https://modelchoice.dev/models/gpt-5-6-sol) | OpenAI | 87.2 | Observed 2026-11-21 · source benchlm:models |
| 15 | [GPT-6 Sol](https://modelchoice.dev/models/gpt-6-sol) | OpenAI | 86.8 | Observed 2026-11-21 · source benchlm:models |
| 16 | [GPT-5.6 Luna](https://modelchoice.dev/models/gpt-5-6-luna) | OpenAI | 84.2 | Observed 2026-11-21 · source benchlm:models |
| 17 | [GPT-5.5](https://modelchoice.dev/models/gpt-5-5) | OpenAI | 78.5 | Observed 2026-11-21 · source benchlm:models |
| 18 | [GPT-6 Luna](https://modelchoice.dev/models/gpt-6-luna) | OpenAI | 71.9 | Observed 2026-11-21 · source benchlm:models |
| 19 | [Beam 1.0 501B-A23B](https://modelchoice.dev/models/reflection-beam) | Reflection AI | 64.7 | 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.
