Qwen3 4B ARC MLX 4bit is an open-source language model by mlx-community. Features: 4b LLM, VRAM: 2GB, Context: 256K, License: apache-2.0, Quantized, LLM Explorer Score: 0.31.
4-bit 4bit Arc Arc-agi Arc-agi-2 Arc-prize Dataset:arc-prize-2026-arc-agi... Mlx Puzzle-solving Qlora Quantized Qwen3 Region:us Safetensors Test-time-training
Model Card on HF 🤗: https://huggingface.co/mlx-community/Qwen3-4B-ARC-MLX-4bit
Qwen3 4B ARC MLX 4bit Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Qwen3 4B ARC MLX 4bit Parameters and Internals
| LLM Name | Qwen3 4B ARC MLX 4bit |
| Repository 🤗 | https://huggingface.co/mlx-community/Qwen3-4B-ARC-MLX-4bit |
| Base Model(s) | |
| Model Size | 4b |
| Required VRAM | 2 GB |
| Updated | 2026-04-10 |
| Maintainer | mlx-community |
| Model Type | qwen3 |
| Model Files | |
| Quantization Type | 4bit |
| Model Architecture | Qwen3ForCausalLM |
| License | apache-2.0 |
| Context Length | 262144 |
| Model Max Length | 262144 |
| Transformers Version | 4.53.3 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 16 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Best Alternatives to Qwen3 4B ARC MLX 4bit
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...Instruct 2507 Unsloth Bnb 4bit | 256K / 3.5 GB | 189920 | 14 |
| Qwen3 4B Self Thinking 16bit | 256K / 8.1 GB | 658 | 0 |
| ...wen3 4B Thinking 2507 MLX 4bit | 256K / 2.3 GB | 77563 | 11 |
| ...wen3 4B Thinking 2507 MLX 8bit | 256K / 4.3 GB | 74153 | 7 |
| Qwen3 4B Instruct 2507 4bit | 256K / 2.3 GB | 42225 | 9 |
| Fact Extractor Dev 1b | 256K / 8.1 GB | 235 | 0 |
| Jan V3 4B Base Instruct 4bit | 256K / 2.3 GB | 750 | 2 |
| ...Thinking 2507 Unsloth Bnb 4bit | 256K / 3.5 GB | 14904 | 2 |
| ...4B Cybersecurity Heretic 16bit | 256K / 8.1 GB | 44 | 2 |
| ...wen3 4B Instruct 2507 Bnb 4bit | 256K / 2.6 GB | 6091 | 4 |
Rank the Qwen3 4B ARC MLX 4bit Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 52721 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a