DeepHermes Tool Calling Specialist
This Atropos release fine-tunes DeepHermes 3 Llama 3.1 8B for reasoning-based tool calls.
Research release
Training and model behavior
The card reports separate simple and parallel function-call evaluations. To combine reasoning and tool use, it instructs users to place the reasoning system prompt before the function-calling prompt. It remains an experimental artifact.DeepHermes Tool Calling Specialist model card
Working with this model
Practical guidance. Evaluate schema compliance and tool selection separately. A correct function name can still carry invalid arguments. Keep tool execution under application control.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Research release |
| Launch date | April 2025 |
| Access | Downloadable |
| Model identifiers | NousResearch/DeepHermes-ToolCalling-Specialist-Atropos |
| Base model | DeepHermes 3 Llama 3.1 8BDeepHermes Tool Calling Specialist model card |
| Repository license label | llama3DeepHermes Tool Calling Specialist model card |
Related models
- DeepHermes Ascension Maze
Nous Research · Atropos experiments
An experimental DeepHermes reinforcement-learning artifact.
Launch date April 2025Research releaseDownloadable - DeepHermes Egregore v1
Nous Research · Atropos experiments
An experimental DeepHermes reinforcement-learning artifact.
Launch date April 2025Research releaseDownloadable - DeepHermes Egregore v2
Nous Research · Atropos experiments
An experimental DeepHermes reinforcement-learning artifact.
Launch date April 2025Research releaseDownloadable