Model Overview
Model Features
Model Capabilities
Use Cases
🚀 Llamacpp imatrix Quantizations of Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf by e-n-v-y
This project offers quantized versions of the Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf
model. It provides various quantization types to meet different performance and storage requirements, enabling users to run the model efficiently on different platforms.
🚀 Quick Start
- Using LM Studio: You can run the quantized models in LM Studio.
- Using llama.cpp: Run the models directly with llama.cpp, or any other llama.cpp based project.
✨ Features
- Multiple Quantization Types: Offers a wide range of quantization types, such as Q8_0, Q6_K, Q5_K_M, etc., to balance between model quality and file size.
- Online Repacking: Some quantization types support online repacking, which can automatically optimize weights for better performance on ARM and AVX machines.
📦 Installation
Prerequisites
Make sure you have huggingface-cli
installed:
pip install -U "huggingface_hub[cli]"
Download a Single File
huggingface-cli download bartowski/e-n-v-y_Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-GGUF --include "e-n-v-y_Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q4_K_M.gguf" --local-dir ./
Download Split Files
If the model is bigger than 50GB and split into multiple files, run:
huggingface-cli download bartowski/e-n-v-y_Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-GGUF --include "e-n-v-y_Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q8_0/*" --local-dir ./
💻 Usage Examples
Prompt Format
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
{system_prompt}<|eot_id|>
<|start_header_id|>user<|end_header_id|>
{prompt}<|eot_id|>
<|start_header_id|>assistant<|end_header_id|>
📚 Documentation
Download a File
Filename | Quant type | File Size | Split | Description |
---|---|---|---|---|
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q8_0.gguf | Q8_0 | 74.98GB | true | Extremely high quality, generally unneeded but max available quant. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q6_K.gguf | Q6_K | 57.89GB | true | Very high quality, near perfect, recommended. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q5_K_M.gguf | Q5_K_M | 49.95GB | true | High quality, recommended. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q5_K_S.gguf | Q5_K_S | 48.66GB | false | High quality, recommended. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q4_1.gguf | Q4_1 | 44.31GB | false | Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q4_K_L.gguf | Q4_K_L | 43.30GB | false | Uses Q8_0 for embed and output weights. Good quality, recommended. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q4_K_M.gguf | Q4_K_M | 42.52GB | false | Good quality, default size for most use cases, recommended. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q4_K_S.gguf | Q4_K_S | 40.35GB | false | Slightly lower quality with more space savings, recommended. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q4_0.gguf | Q4_0 | 40.12GB | false | Legacy format, offers online repacking for ARM and AVX CPU inference. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ4_NL.gguf | IQ4_NL | 40.05GB | false | Similar to IQ4_XS, but slightly larger. Offers online repacking for ARM CPU inference. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q3_K_XL.gguf | Q3_K_XL | 38.06GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ4_XS.gguf | IQ4_XS | 37.90GB | false | Decent quality, smaller than Q4_K_S with similar performance, recommended. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q3_K_L.gguf | Q3_K_L | 37.14GB | false | Lower quality but usable, good for low RAM availability. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q3_K_M.gguf | Q3_K_M | 34.27GB | false | Low quality. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ3_M.gguf | IQ3_M | 31.94GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q3_K_S.gguf | Q3_K_S | 30.91GB | false | Low quality, not recommended. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ3_XS.gguf | IQ3_XS | 29.31GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ3_XXS.gguf | IQ3_XXS | 27.47GB | false | Lower quality, new method with decent performance, comparable to Q3 quants. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q2_K_L.gguf | Q2_K_L | 27.40GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-Q2_K.gguf | Q2_K | 26.38GB | false | Very low quality but surprisingly usable. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ2_M.gguf | IQ2_M | 24.12GB | false | Relatively low quality, uses SOTA techniques to be surprisingly usable. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ2_S.gguf | IQ2_S | 22.24GB | false | Low quality, uses SOTA techniques to be usable. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ2_XS.gguf | IQ2_XS | 21.14GB | false | Low quality, uses SOTA techniques to be usable. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ2_XXS.gguf | IQ2_XXS | 19.10GB | false | Very low quality, uses SOTA techniques to be usable. |
Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf-IQ1_M.gguf | IQ1_M | 16.75GB | false | Extremely low quality, not recommended. |
Embed/output weights
Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.
ARM/AVX information
Previously, you would download Q4_0_4_4/4_8/8_8, and these would have their weights interleaved in memory in order to improve performance on ARM and AVX machines by loading up more data in one pass.
Now, however, there is something called "online repacking" for weights. details in this PR. If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.
As of llama.cpp build b4282 you will not be able to run the Q4_0_X_X files and will instead need to use Q4_0.
Additionally, if you want to get slightly better quality for , you can use IQ4_NL thanks to this PR which will also repack the weights for ARM, though only the 4_4 for now. The loading time may be slower but it will result in an overall speed incrase.
Click to view Q4_0_X_X information (deprecated
I'm keeping this section to show the potential theoretical uplift in performance from using the Q4_0 with online repacking.
Click to view benchmarks on an AVX2 system (EPYC7702)
model | size | params | backend | threads | test | t/s | % (vs Q4_0) |
---|---|---|---|---|---|---|---|
qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | pp512 | 204.03 ± 1.03 | 100% |
qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | pp1024 | 282.92 ± 0.19 | 100% |
qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | pp2048 | 259.49 ± 0.44 | 100% |
qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | tg128 | 39.12 ± 0.27 | 100% |
qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | tg256 | 39.31 ± 0.69 | 100% |
qwen2 3B Q4_0 | 1.70 GiB | 3.09 B | CPU | 64 | tg512 | 40.52 ± 0.03 | 100% |
qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | pp512 | 301.02 ± 1.74 | 147% |
qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | pp1024 | 287.23 ± 0.20 | 101% |
qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | pp2048 | 262.77 ± 1.81 | 101% |
qwen2 3B Q4_K_M | 1.79 GiB | 3.09 B | CPU | 64 | tg128 | 18.80 ± 0.99 |
🔧 Technical Details
- Quantization Method: Using llama.cpp release b5524 for quantization.
- Original Model: The original model can be found at https://huggingface.co/e-n-v-y/Legion-V2.1-LLaMa-70B-Elarablated-v0.8-hf.
- Quantization Dataset: All quants are made using the imatrix option with the dataset from here.
📄 License
This project is licensed under the llama3.3
license.

