Hermes Agent
Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.
Model usage, over time.
Daily and weekly source totals, grouped into GPTProto catalog models.
Usage grouped into GPTProto catalog models. Unlisted source models are reassigned using the existing catalog rules; these are not the reference platform’s original per-model rankings.
The final day/week may be incomplete at the snapshot time. No full-period forecast is included.
8 catalog models after grouping; 20 public source rows; 548 models reported by the source.
The models behindHermes Agent.
Usage grouped into GPTProto catalog models. Unlisted source models are reassigned using the existing catalog rules; these are not the reference platform’s original per-model rankings.
GPTProto model allocation for Hermes Agent
GPTProto prices where available. — means unavailable, not free.
| Model | Price: GPTProto | vs Official | Tokens |
|---|---|---|---|
1.  GPT 6.1 TerraOpenAI | $0.80 / $4.00$0.04 Cache read · per 1M tokens | −20% | 32.7T |
2.  DeepSeek v4 FlashDeepSeek | $0.24 / $0.96$0.00 Cache read · per 1M tokens | −20% | 4.5T |
3.  GLM 5.3 FlashZ-AI | $0.13 / $0.42$0.03 Cache read · per 1M tokens | −15% | 4T |
4. GPT 6 LunaOpenAI | $0.080 / $0.40$0.01 Cache read · per 1M tokens | −20% | 1.6T |
5. GLM 5.3Z-AI | $1.19 / $3.74$0.22 Cache read · per 1M tokens | −15% | 828B |
6. DeepSeek v4 ProDeepSeek | $1.06 / $3.17$0.04 Cache read · per 1M tokens | −20% | 753B |
7.  Gemini 3.8 FlashGoogle | $0.90 / $4.50$0.09 Cache read · per 1M tokens | −40% | 602B |
8.  Minimax M3MiniMax | $0.48 / $0.96$0.10 Cache read · per 1M tokens | −20% | 599B |