4-bit Autotrain compatible Awq Custom code Endpoints compatible Ext 8k Mpt Pytorch Quantized Region:us
Model Card on HF 🤗: https://huggingface.co/casperhansen/mpt-7b-8k-chat-awq
Mpt 7B 8K Chat AWQ Benchmarks
ARC: 48.04 vs 96.7 (so35)-50.3%
HellaSwag: 77.62 vs 95.3 (gpt4)-18.6%
MMLU: 41.88 vs 88.3 (so35)-52.6%
TruthfulQA: 43.68 vs 59 (gpt4)-26%
WinoGrande: 71.03 vs 87.5 (gpt4)-18.8%
GSM8K: 4.4 vs 96.4 (so35)-95.4%
LLME Score: 0.12748
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)
Mpt 7B 8K Chat AWQ Parameters and Internals
| LLM Name | Mpt 7B 8K Chat AWQ |
| Repository 🤗 | https://huggingface.co/casperhansen/mpt-7b-8k-chat-awq |
| Base Model(s) | |
| Model Size | 7b |
| Required VRAM | 3.8 GB |
| Updated | 2025-11-15 |
| Maintainer | casperhansen |
| Model Type | mpt |
| Model Files | |
| AWQ Quantization | Yes |
| Context Length | 8k |
| Quantization Type | awq |
| Model Architecture | MPTForCausalLM |
| Model Max Length | 8192 |
| Transformers Version | 4.30.2 |
| Tokenizer Class | GPTNeoXTokenizer |
| Vocabulary Size | 50432 |
| Torch Data Type | bfloat16 |
Best Alternatives to Mpt 7B 8K Chat AWQ
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Mpt 7B AWQ | 0K / 3.8 GB | 12 | 0 |
| Mpt 7B Chat 8K 4bit AWQ | 0K / 3.8 GB | 13 | 1 |
| ...saicml Mpt 7B Chat W4 G128 AWQ | 0K / 3.8 GB | 21 | 0 |
| ...ml Mpt 7B Instruct W4 G128 AWQ | 0K / 3.8 GB | 10 | 0 |
| Mpt 7B Chat Q8 | 0K / 6.9 GB | 21 | 1 |
| Mpt 7B Storywriter Q8 | 0K / 6.9 GB | 22 | 2 |
| Mpt 7B Instruct Q8 | 0K / 6.9 GB | 0 | 2 |
| Mpt 7B Q8 | 0K / 6.9 GB | 1 | 1 |
| Mpt 7B Storywriter 4bit 128g | 0K / 3.9 GB | 25 | 125 |
| ...writer 4bit 128g 65kTokens CPU | 0K / 3.9 GB | 0 | 9 |
Note: green Score (e.g. "73.2") means that the model is better than casperhansen/mpt-7b-8k-chat-awq .
Rank the Mpt 7B 8K Chat AWQ 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? 51557 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124