8-bit 8bit Autotrain compatible Custom code Endpoints compatible Pytorch Quantized Refinedweb Region:us Sharded
Model Card on HF 🤗: https://huggingface.co/substratusai/falcon-40b-8bit
Falcon 40B 8bit 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)
Falcon 40B 8bit Parameters and Internals
| LLM Name | Falcon 40B 8bit |
| Repository 🤗 | https://huggingface.co/substratusai/falcon-40b-8bit |
| Model Size | 40b |
| Required VRAM | 41.8 GB |
| Updated | 2025-08-21 |
| Maintainer | substratusai |
| Model Type | RefinedWeb |
| Model Files | |
| Quantization Type | 8bit |
| Model Architecture | RWForCausalLM |
| Model Max Length | 2048 |
| Transformers Version | 4.31.0.dev0 |
| Is Biased | 0 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Vocabulary Size | 65024 |
| Torch Data Type | float16 |
Best Alternatives to Falcon 40B 8bit
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Falcon 40B Instruct 8bit | 0K / 41.8 GB | 13 | 6 |
| Alfred 40B 1023 | 0K / 83.6 GB | 2120 | 48 |
| Vulture 40B | 0K / 81.8 GB | 1926 | 8 |
| Alfred 40B 1023 AWQ | 0K / 23.3 GB | 13 | 5 |
| Alfred 40B 1023 GPTQ | 0K / 22.5 GB | 14 | 3 |
| FalconLite | 0K / 22.3 GB | 349 | 170 |
| FalconLite | 0K / 22.3 GB | 284 | 170 |
| Docsgpt 40B Falcon | 0K / 82.5 GB | 28 | 13 |
| Alfred 40B 0723 | 0K / 83.6 GB | 24 | 46 |
| Falcon 40B Instruct GPTQ | 0K / 22.5 GB | 323 | 197 |
Note: green Score (e.g. "73.2") means that the model is better than substratusai/falcon-40b-8bit .
Rank the Falcon 40B 8bit 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? 50804 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124