A 70-billion parameter instruction-tuned language model from Nous with a 131K token context window.
- Strengths
- Handles complex multi-turn conversations and agentic reasoning tasks with a large context window for long-form inputs.
- Best for
- Applications requiring extended context and instruction-following at the 70B parameter scale, before Hermes 4 70B superseded it.
- Limitations
- Replaced by Hermes 4 70B (August 2025), which offers improvements at the same parameter and context size; smaller than Hermes 3 405B Instruct for tasks demanding maximum capacity.
Input / 1M
$0.7
Output / 1M
$0.7
Cached input / 1M
—
Context window
131K
Price history
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 11 Jun 2026 | $0.7 | $0.7 | — | Imported from OpenRouter | openrouter.ai |