Bespoke-Nimble-9B
Bespoke-Nimble-9B is a Qwen3.5-9B LoRA adapter for context-grounded choices, booleans, and rubric score levels. It scores allowed answer tokens directly rather than generating reasoning or free-form answers.
What it does
Use it to extract structured choices, boolean decisions, or rubric scores from supplied context.
- Availability
- Adapter, tokenizer, prompt builder, and reference inference code are available in the Hugging Face repository.
- Pricing model
- Not established
- Deployment
- Requires a CUDA GPU with BF16 support; Mac/MLX and Linux loaders are described in the GitHub quickstart.
- Privacy
- Not established
- Licence
- Apache 2.0
Capability
Modality
Access
Deployment
Pricing
Target user
Evidence
6 claims- The adapter supports context-grounded choices, booleans, and rubric score levels. [1]
A Qwen3.5-9B LoRA adapter for context-grounded choices, booleans, and rubric score levels.
Bespoke Labs. Quoted text is outside CC BY 4.0 and keeps its original rights. - Each field supports at most 255 choices. [1]
Each field supports at most 255 choices.
Bespoke Labs. Quoted text is outside CC BY 4.0 and keeps its original rights. - Prompts exceeding 8,192 tokens are rejected rather than truncated. [1]
Prompts exceeding 8,192 tokens are rejected rather than truncated.
Bespoke Labs. Quoted text is outside CC BY 4.0 and keeps its original rights. - The repository contains the adapter, tokenizer, exact prompt builder, reference inference code, and release metadata. [1]
This repository contains the adapter (about 165 MiB), tokenizer, exact prompt builder, reference inference code, and release metadata.
Bespoke Labs. Quoted text is outside CC BY 4.0 and keeps its original rights. - The adapter requires the Qwen3.5-9B base checkpoint. [1]
It requires the [Qwen3.5-9B base checkpoint](https://huggingface.co/Qwen/Qwen3.5-9B/tree/c202236235762e1c871ad0ccb60c8ee5ba337b9a).
Bespoke Labs. Quoted text is outside CC BY 4.0 and keeps its original rights. - The release uses T=1.0, and the earlier checkpoint remains available at revision `original-2676`. [1]
This release uses T=1.0; the earlier checkpoint remains available at `revision="original-2676"`, and [v2](https://huggingface.co/bespokelabs/Bespoke-Nimble-9B-v2) is unchanged.
Bespoke Labs. Quoted text is outside CC BY 4.0 and keeps its original rights.
Sources
1
1
Bespoke LabsOwner-controlled
https://huggingface.co/bespokelabs/Bespoke-Nimble-9B
Retrieved 2026-10-27 01:28 UTC
Verified
Versions
3| Version | Effective | Change | Changed fields | State |
|---|---|---|---|---|
| 3 | 2026-10-18 15:45 UTC | evidence update | score | Current |
| 2 | 2026-09-25 13:23 UTC | material change | availability, deployment | Current |
| 1 | 2026-09-24 13:42 UTC | publication | None | Current |
In briefs
2- New models and tools focus on typed decisions and long-context tasks Weekly brief 2026-W39
- Claude Platform adds automatic evaluation for Managed Agents tool calls Daily brief 2026-09-24
Original catalog content is licensed under CC BY 4.0, attributed to wowsuchai. Source quotations and linked content keep their original rights.