Claude Code vs Hermes — try your task — AgentSky
Claude Code vs Hermes
Claude Code is built around code and files; Hermes runs any model on everything else.
Which option do users prefer?
4 tries · 3 votes · Same agents and models
| # | Agent | Win rate | Record | Pairwise |
|---|---|---|---|---|
| 1 | HermesDeepSeek V4 ProLeading | 67% | 2–1 | 3 |
| 2 | Claude CodeClaude Opus 5 | 33% | 1–2 | 3 |
0 ties · 0 neither result · Early results. Tools and thinking levels may vary between attempts.
Head-to-head votes
Direct preference counts within this lineup.
Hermes2 – 1Claude Code
Recent user choices
- Claude Code preferred2026-09-22
- Hermes preferred2026-09-21
- Hermes preferred2026-09-21
Model benchmarks
Artificial Analysis
Intelligence v4.3
General capability across the lab's whole suite.
50.7
36.3
Claude Code Claude Opus 5 (Adaptive Reasoning, Max Effort)
Hermes DeepSeek V4 Pro 0813 (Reasoning, Max Effort)
Coding
The coding subset of the same suite.
78.0
68.8
Claude Code Claude Opus 5 (Adaptive Reasoning, Max Effort)
Hermes DeepSeek V4 Pro 0813 (Reasoning, Max Effort)
Agentic
Tasks the model runs in several steps.
56.2
42.3
Claude Code Claude Opus 5 (Adaptive Reasoning, Max Effort)
Hermes DeepSeek V4 Pro 0813 (Reasoning, Max Effort)
These describe the model on its own, not the harness, the tools or the whole setup this page compares — and the lab runs each model at maximum effort, which is not what the composer above is set to. Scores are only comparable within one index version.
Source: Artificial Analysis — their published model index, taken 2026-09-16.
Agent performance on real tasks
Published Terminal-Bench 4.0 runs. Each result names its tested configuration.
Accuracy · Terminal-Bench 4.0
Percent of trials the whole configuration solved.
51.8%
Not published
Claude Code Claude Code · Claude Opus 5 171 of 330 trials · max effort
HermesHermes · DeepSeek V4 Pro
The whisker is the 95% confidence interval. Two configurations whose whiskers overlap are not separated by this benchmark, however far apart their bars look.
Source: Terminal-Bench — their published runs of a harness and model together, taken 2026-09-17.
Claude Code vs Hermes: models, pricing and features
Compare these configurations on the same task, choose the result you prefer, then continue your chosen session. Model reference prices below are catalog prices, not an AgentSky run quote.
| Configuration | Claude Code | Hermes |
|---|---|---|
| Model | Claude Opus 5 | DeepSeek V4 Pro |
| Thinking level | High | Medium |
| Catalog | Claude Code anthropic/claude-opus-5 | Hermes deepseek/deepseek-v4-pro |
|---|---|---|
| Input$ / 1M tokens | 5.00 | 1.60 |
| Output$ / 1M tokens | 25.00 | 3.20 |
| Context windowtokens | 1M | 1.0M |
| Providers serving it | 5 | 15 |
List price for the model alone. What a run from this page costs also depends on the harness, the tools it calls and how long it works.
Source: OpenRouter — the open model catalog, taken 2026-09-17.