36 lines
883 B
Markdown
36 lines
883 B
Markdown
|
---
|
||
|
|
tags:
|
||
|
|
- text-generation-inference
|
||
|
|
- transformers
|
||
|
|
- unsloth
|
||
|
|
- qwen3
|
||
|
|
license: apache-2.0
|
||
|
|
datasets:
|
||
|
|
- TeichAI/gemini-3-pro-preview-high-reasoning-1000x
|
||
|
|
base_model:
|
||
|
|
- unsloth/Qwen3-4B-Instruct-2507
|
||
|
|
---
|
||
|
|
|
||
|
|
# Qwen3 4B Instruct 2507 - Gemini 3 Pro Preview (No Reasoning) Distill
|
||
|
|
|
||
|
|
This model was trained on a **Gemini 3 Pro Preview** dataset with a high reasoning effort.
|
||
|
|
|
||
|
|
The reasoning summaries were then formatted out of the dataset and the model was finetuned on the final answers only.
|
||
|
|
|
||
|
|
- 🧬 Datasets:
|
||
|
|
- `TeichAI/gemini-3-pro-preview-high-reasoning-1000x`
|
||
|
|
|
||
|
|
- 🏗 Base Model:
|
||
|
|
- `unsloth/Qwen3-4B-Instruct-2507`
|
||
|
|
|
||
|
|
- ⚡ Use cases:
|
||
|
|
- Coding
|
||
|
|
- Science
|
||
|
|
|
||
|
|
- ∑ Stats (Dataset)
|
||
|
|
- Costs: $ 32.7 (USD)
|
||
|
|
- Total tokens (input + output): 2.73 M
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
This qwen3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|