--- duplicated_from: localmodels/LLM --- # Vicuna 13B v1.3 ggml From LMSYS: https://huggingface.co/lmsys/vicuna-13b-v1.3 --- ### Original llama.cpp quant methods: `q4_0, q4_1, q5_0, q5_1, q8_0` Quantized using an older version of llama.cpp and compatible with llama.cpp from May 19, commit 2d5db48. ### k-quant methods: `q2_K, q3_K_S, q3_K_M, q3_K_L, q4_K_S, q4_K_M, q5_K_S, q6_K` Quantization methods compatible with latest llama.cpp from June 6, commit 2d43387. --- ## Files | Name | Quant method | Bits | Size | Max RAM required, no GPU offloading | Use case | | ---- | ---- | ---- | ---- | ---- | ----- | | vicuna-13b-v1.3.0.ggmlv3.q2_K.bin | q2_K | 2 | 5.51 GB | 8.01 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.vw and feed_forward.w2 tensors, GGML_TYPE_Q2_K for the other tensors. | | vicuna-13b-v1.3.0.ggmlv3.q3_K_L.bin | q3_K_L | 3 | 6.93 GB | 9.43 GB | New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K | | vicuna-13b-v1.3.0.ggmlv3.q3_K_M.bin | q3_K_M | 3 | 6.31 GB | 8.81 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K | | vicuna-13b-v1.3.0.ggmlv3.q3_K_S.bin | q3_K_S | 3 | 5.66 GB | 8.16 GB | New k-quant method. Uses GGML_TYPE_Q3_K for all tensors | | vicuna-13b-v1.3.0.ggmlv3.q4_0.bin | q4_0 | 4 | 7.32 GB | 9.82 GB | Original llama.cpp quant method, 4-bit. | | vicuna-13b-v1.3.0.ggmlv3.q4_1.bin | q4_1 | 4 | 8.14 GB | 10.64 GB | Original llama.cpp quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. | | vicuna-13b-v1.3.0.ggmlv3.q4_K_M.bin | q4_K_M | 4 | 7.87 GB | 10.37 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K | | vicuna-13b-v1.3.0.ggmlv3.q4_K_S.bin | q4_K_S | 4 | 7.37 GB | 9.87 GB | New k-quant method. Uses GGML_TYPE_Q4_K for all tensors | | vicuna-13b-v1.3.0.ggmlv3.q5_0.bin | q5_0 | 5 | 8.95 GB | 11.45 GB | Original llama.cpp quant method, 5-bit. Higher accuracy, higher resource usage and slower inference. | | vicuna-13b-v1.3.0.ggmlv3.q5_1.bin | q5_1 | 5 | 9.76 GB | 12.26 GB | Original llama.cpp quant method, 5-bit. Even higher accuracy, resource usage and slower inference. | | vicuna-13b-v1.3.0.ggmlv3.q5_K_M.bin | q5_K_M | 5 | 9.23 GB | 11.73 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K | | vicuna-13b-v1.3.0.ggmlv3.q5_K_S.bin | q5_K_S | 5 | 8.97 GB | 11.47 GB | New k-quant method. Uses GGML_TYPE_Q5_K for all tensors | | vicuna-13b-v1.3.0.ggmlv3.q6_K.bin | q6_K | 6 | 10.68 GB | 13.18 GB | New k-quant method. Uses GGML_TYPE_Q8_K - 6-bit quantization - for all tensors | | vicuna-13b-v1.3.0.ggmlv3.q8_0.bin | q8_0 | 8 | 13.83 GB | 16.33 GB | Original llama.cpp quant method, 8-bit. Almost indistinguishable from float16. High resource use and slow. Not recommended for most users. | --- # Vicuna Model Card ## Model Details Vicuna is a chat assistant trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT. - **Developed by:** [LMSYS](https://lmsys.org/) - **Model type:** An auto-regressive language model based on the transformer architecture. - **License:** Non-commercial license - **Finetuned from model:** [LLaMA](https://arxiv.org/abs/2302.13971). ### Model Sources - **Repository:** https://github.com/lm-sys/FastChat - **Blog:** https://lmsys.org/blog/2023-03-30-vicuna/ - **Paper:** https://arxiv.org/abs/2306.05685 - **Demo:** https://chat.lmsys.org/ ## Uses The primary use of Vicuna is research on large language models and chatbots. The primary intended users of the model are researchers and hobbyists in natural language processing, machine learning, and artificial intelligence. ## How to Get Started with the Model Command line interface: https://github.com/lm-sys/FastChat#vicuna-weights. APIs (OpenAI API, Huggingface API): https://github.com/lm-sys/FastChat/tree/main#api. ## Training Details Vicuna v1.3 is fine-tuned from LLaMA with supervised instruction fine-tuning. The training data is around 140K conversations collected from ShareGPT.com. See more details in the "Training Details of Vicuna Models" section in the appendix of this [paper](https://arxiv.org/pdf/2306.05685.pdf). ## Evaluation Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this [paper](https://arxiv.org/pdf/2306.05685.pdf). ## Difference between different versions of Vicuna See [vicuna_weights_version.md](https://github.com/lm-sys/FastChat/blob/main/docs/vicuna_weights_version.md)