# Hermes Index 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-hermes-index

Hermes Index: agentic capability and cost per task inside Hermes Agent. Every model the catalog carries a published Hermes Index value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Agentic |
| Measure | Mean pass@1 score across four suites, with mean cost per task |
| Tasks | Four suites: Hermes Bench (150 tasks), TerminalBench 4 without its 4 GPU tasks, TerminalBench Science and SkillsBench |
| Difficulty | Frontier agentic |
| Published by | [Hermes Index: an agentic model leaderboard](https://portal.nousresearch.com/bench) |

An index of agentic capability from Nous Research. It averages pass@1 scores on four suites, Hermes Bench, TerminalBench 4, TerminalBench Science and SkillsBench, with every run inside the Hermes Agent harness, and reports the mean cost per task beside the score.

## Hermes Index ranking

17 models with a published Hermes Index 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 | Mean pass@1 score across four suites, with mean cost per task | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Claude Opus 5.5](https://modelscale.dev/models/claude-opus-5-5) | Anthropic | 63.3 | Observed 2026-10-23 · source benchlm:models |
| 2 | [GPT-6.1 Sol](https://modelscale.dev/models/gpt-6-1-sol) | OpenAI | 57.1 | Observed 2026-10-23 · source benchlm:models |
| 3 | [GPT-6 Astra](https://modelscale.dev/models/gpt-6-astra) | OpenAI | 56.3 | Observed 2026-10-23 · source benchlm:models |
| 4 | [Claude Sonnet 5.5](https://modelscale.dev/models/claude-sonnet-5-5) | Anthropic | 53.1 | Observed 2026-10-23 · source benchlm:models |
| 5 | [GPT-6 Sol](https://modelscale.dev/models/gpt-6-sol) | OpenAI | 44.1 | Observed 2026-10-23 · source benchlm:models |
| 6 | [Grok 4.7](https://modelscale.dev/models/grok-4-7) | xAI | 39.3 | Observed 2026-10-23 · source benchlm:models |
| 7 | [GLM-5.3](https://modelscale.dev/models/glm-5-3) | Z.AI | 39.2 | Observed 2026-10-23 · source benchlm:models |
| 8 | [Kimi K3](https://modelscale.dev/models/kimi-3) | Moonshot AI | 39.2 | Observed 2026-10-23 · source benchlm:models |
| 9 | [Gemini 3.8 Flash](https://modelscale.dev/models/gemini-3-8-flash) | Google | 38.1 | Observed 2026-10-23 · source benchlm:models |
| 10 | [Claude Haiku 5.5](https://modelscale.dev/models/claude-haiku-5-5) | Anthropic | 37.0 | Observed 2026-10-23 · source benchlm:models |
| 11 | [DeepSeek V4.1 Flash 1017](https://modelscale.dev/models/deepseek-v4-1-flash) | DeepSeek | 36.9 | Last good 2026-10-19 · source benchlm:models |
| 12 | [Qwen3.8 Max](https://modelscale.dev/models/qwen3-8-max) | Alibaba | 36.2 | Observed 2026-10-23 · source benchlm:models |
| 13 | [GLM-5.3-Flash](https://modelscale.dev/models/glm-5-3-flash) | Z.AI | 35.0 | Observed 2026-10-23 · source benchlm:models |
| 14 | [GPT-6 Luna](https://modelscale.dev/models/gpt-6-luna) | OpenAI | 33.9 | Observed 2026-10-23 · source benchlm:models |
| 15 | [Step 5 Preview](https://modelscale.dev/models/step-5-preview) | StepFun | 32.8 | Observed 2026-10-23 · source benchlm:models |
| 16 | [Hy4 preview](https://modelscale.dev/models/hy4-preview) | Tencent | 26.1 | Observed 2026-10-23 · source benchlm:models |
| 17 | [Ling 3.0 Flash](https://modelscale.dev/models/ling-3-0-flash) | InclusionAI | 21.6 | Observed 2026-10-23 · 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.