Downloads · 30 days
35
4% of all-time downloads
Qwen3-4B-Instruct-NoThink-V2.1
VladHong/Qwen3-4B-Instruct-NoThink-V2.1
Qwen3-4B-Instruct-NoThink-V2.1 is a machine learning model from VladHong. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Finetuned from Unsloth/Qwen3-4B-Instruct-2507 using QLoRA + Unsloth. Trained to respond directly without chain-of-thought (<think blocks stripped from all training data).
All-time downloads
917
Public
Repo size
4.8 GB
Likes
2
Public
Repo makeup
Click a slice to open those files.
.gguf2.3 GB · 94%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Mar 3, 2026
- Updated
- Mar 8, 2026
- SHA
- edfaa4c9
Base models
Datasets
- TeichAI/gemini-3-pro-preview-high-reasoning-250x
- TeichAI/gemini-3-pro-preview-high-reasoning-1000x
- Roman1111111/gemini-3.1-pro-hard-high-reasoning
- crownelius/Opus-4.6-Reasoning-3300x
- TeichAI/claude-4.5-opus-high-reasoning-250x
- TeichAI/claude-sonnet-4.5-high-reasoning-250x
- TeichAI/gpt-5.2-high-reasoning-250x
Model card
From the Hugging Face model README
Qwen3-4B Instruct No-Think V2.1
Finetuned from Unsloth/Qwen3-4B-Instruct-2507 using QLoRA + Unsloth. Trained to respond directly without chain-of-thought (<think> blocks stripped from all training data).
What's New in V2.1: The best I can afford on free Kaggle GPU
V2.1 further expands the training data to 7 datasets covering Gemini, Claude, and GPT-sourced reasoning traces, and was trained on Kaggle's dual T4 GPU environment (though only one GPU was used for training). Dataset size after deduplication contains ~4,752 examples with MinHash-based similarity filtering at 0.4 Jaccard threshold.
Training Data
| Dataset | Source Model |
|---|---|
| TeichAI/gemini-3-pro-preview-high-reasoning-250x | Gemini |
| TeichAI/gemini-3-pro-preview-high-reasoning-1000x | Gemini |
| Roman1111111/gemini-3.1-pro-hard-high-reasoning | Gemini |
| crownelius/Opus-4.6-Reasoning-3300x | Claude |
| TeichAI/claude-4.5-opus-high-reasoning-250x | Claude |
| TeichAI/claude-sonnet-4.5-high-reasoning-250x | Claude |
| TeichAI/gpt-5.2-high-reasoning-250x | GPT |
~7,322 rows loaded, ~4,752 after MinHash deduplication (threshold 0.4). <think> blocks stripped from all assistant turns before training.
Training Details
| Parameter | Value |
|---|---|
| Method | QLoRA (4-bit NF4) + Unsloth |
| LoRA rank | 16 |
| LoRA alpha | 16 |
| Epochs | 1 |
| Steps | 594 |
| Batch size | 2 per device × 4 gradient accumulation = 8 effective |
| Learning rate | 1e-4 (cosine schedule) |
| Max seq length | 2048 |
| Optimizer | AdamW 8-bit |
| Hardware | Kaggle Tesla T4 (14.56 GB VRAM) |
| Training time | ~7h 23min |
| Trainable params | 33M / 4.05B (0.81%) |
| Peak VRAM | ~10.2 GB |
Training used train_on_responses_only — loss computed on assistant completions only.
Files
*.gguf— IQ4_XS quantized, ready for LM Studio / Ollama / llama.cpplora-adapter/— Raw LoRA weights for merging with the base model
Usage (Ollama)
ollama run VladHong/Qwen3-4B-Instruct-NoThink-V2.1
License Note
Base model is Apache 2.0. Training data includes AI-generated content — review upstream dataset terms before commercial use.
- License
- apache-2.0
- Languages
- en
- Created
- Mar 3, 2026
- Updated
- Mar 8, 2026