Model Card on HF 🤗: https://huggingface.co/mlx-community/Qwen1.5-14B-Chat-4bit
Qwen1.5 14B Chat 4bit Benchmarks
ARC: 58.7 vs 96.7 (so35)-39.3%
HellaSwag: 82.27 vs 95.3 (gpt4)-13.7%
MMLU: 68.57 vs 88.3 (so35)-22.3%
TruthfulQA: 60.36 vs 59 (gpt4)2.3%
WinoGrande: 73.09 vs 87.5 (gpt4)-16.5%
GSM8K: 30.63 vs 96.4 (so35)-68.2%
LLME Score: 0.17524
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)
Qwen1.5 14B Chat 4bit Parameters and Internals
| Model Type |
| |
| Additional Notes |
|
| LLM Name | Qwen1.5 14B Chat 4bit |
| Repository 🤗 | https://huggingface.co/mlx-community/Qwen1.5-14B-Chat-4bit |
| Base Model(s) | |
| Model Size | 14b |
| Required VRAM | 9.1 GB |
| Updated | 2025-05-17 |
| Maintainer | mlx-community |
| Model Type | qwen2 |
| Model Files | |
| Supported Languages | en |
| Quantization Type | 4bit |
| Model Architecture | Qwen2ForCausalLM |
| License | other |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.38.2 |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 152064 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Best Alternatives to Qwen1.5 14B Chat 4bit
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...B Instruct 1M Unsloth Bnb 4bit | 986K / 14.3 GB | 250 | 3 |
| Qwen2.5 14B Instruct 1M 8bit | 986K / 15.7 GB | 103 | 7 |
| Qwen2.5 14B Instruct 1M 4bit | 986K / 8.3 GB | 104 | 2 |
| ...en2.5 14B Instruct 1M Bnb 4bit | 986K / 9.9 GB | 15 | 2 |
| QwQ 14B Math V0.2 | 128K / 29.7 GB | 147 | 15 |
| ...till Qwen 14B Unsloth Bnb 4bit | 128K / 14.2 GB | 13049 | 29 |
| Qwen Carpmuscle V0.3 | 128K / 29.7 GB | 5 | 0 |
| Qwen2.5 14B Unsloth Bnb 4bit | 128K / 12.1 GB | 4332 | 1 |
| ...k R1 Distill Qwen 14B Bnb 4bit | 128K / 9.9 GB | 3680 | 4 |
| DeepCoder 14B Preview 4bit | 128K / 8.3 GB | 325 | 2 |
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/Qwen1.5-14B-Chat-4bit .
Rank the Qwen1.5 14B Chat 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? 47402 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227