Hermes Index adds GPT 6.1 Sol, now ranks 18 models on agentic tasks; Claude Opus 5.5 tops at 63.31
Teknium · x · 2026-10-18
Teknium highlighted the updated Hermes Index, which now benchmarks 18 models across four suites (Hermes Bench, TerminalBench 4, TerminalBench Science, SkillsBench) inside Hermes Agent, scoring capability vs. cost on real agentic tasks.
Key results:
- Claude Opus 5.5 leads at 63.31 ($4.99/task)
- GPT 6.1 Sol scores 57.09 at just $1.19/task
- GPT 6 Astra (56.25), Claude Sonnet 5.5 (53.14), and GPT 6 Sol (44.10) round out the top five
- Ling 3.0 Flash is cheapest at $0.054/task but scores only 21.56
Hermes connects to most AI subscriptions and providers (Nous Portal, ChatGPT, Claude, OpenRouter, and more), letting users pick models via the Index.
Related event: Hermes Index Updates: Opus 5.5 Tops 18-Model Agent Leaderboard(2 posts)→
More from coding & agent
- Indie maker's fully AI-agent-run YouTube channel notches first video past 1K views — tibo_maker · 2026-10-18
- Luma adds Send to Luma inside Claude Motion, turning animated explainers into real video projects — lmoroney · 2026-10-18
- Small businesses don't need agents: define the workflow first, pick the tool second — DavidLinthicum · 2026-10-18
- grill-me: Matt Pocock's open-source Claude Skill that grills your design decisions — paulabartabajo_ · 2026-10-18
- mitsuhiko: Claude Opus traces macOS BTLEServer CPU spin to an AirPods firmware check bug — mitsuhiko · 2026-10-18
- An Agent Can Remember Perfectly and Still Be Wrong: What Evidence Proves 'Done' After a Context Reset? — jylusdev · 2026-10-18