qwen3-harness-finetune is an open-weight model from Satoshi Sasahara. It has 262,144-token context. Its published files total 6.7 GB.
This model was finetuned and converted to GGUF format using Unsloth. - Qwen3.5-9B.Q4KM.gguf - Qwen3.5-9B.F16-mmproj.gguf This was trained 2x faster with Unsloth
Model Card
This model was finetuned and converted to GGUF format using Unsloth. - Qwen3.5-9B.Q4KM.gguf - Qwen3.5-9B.F16-mmproj.gguf This was trained 2x faster with Unsloth
Excerpt from the card by Satoshi Sasahara.
Configuration
- Architecture
- Qwen3_5ForConditionalGeneration
- Context length (tokens)
- 262,144
- Layers
- 32
- Hidden size
- 4,096
- Feed-forward size
- 12,288
- Attention heads
- 16
- Key/value heads
- 4
- Head dimension
- 256
- Vocabulary size
- 248,320
- Stored precision
- bfloat16
- Model type
- qwen3_5
Identity and Version
- Repository
- papandada/qwen3-harness-finetune
- Publisher
- Satoshi Sasahara
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- e949d1df029eb30d6c56bdc62c52c468fab6b8c2
- First published
- 2026-09-19
- Last updated
- 2026-09-19
Files and Weights
5 files, 6.7 GB in total. The weights are 2 files totalling 6.7 GB in gguf.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| Qwen3.5-9B.F16-mmproj.gguf | Weights | 918.2 MB | 5cff72327518 |
| Qwen3.5-9B.Q4_K_M.gguf | Weights | 5.8 GB | 74f39be7f926 |
| config.json | Configuration | 2.8 KB | — |
| README.md | Documentation | 726 B | — |
| .gitattributes | Repository | 1.6 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 6.7 GB
Released by Satoshi Sasahara through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 6.7 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About qwen3-harness-finetune
What is qwen3-harness-finetune's context length?
262,144 tokens, from the maximum position embeddings in its published configuration.