lmsys

--- inference: false license: llama2 ---

This model is copied from https://huggingface.co/yuhuili/EAGLE-llama2-chat-7B with the following modifications to make it compatible with SGLang: - Modify the architecture in config.json to LlamaFo...

gpt-oss-20b-bf16 Model Introduction This model is the bf16 version converted from openai/gpt-oss-20b. Usage You can use this model in SGLang with the following instructions. Installation For more details https://github.com/sgl-project/sglang/issues/8833

gpt-oss-120b-bf16 Model Introduction This model is the bf16 version converted from openai/gpt-oss-120b. Usage You can use this model in SGLang with the following instructions. Installation For more details https://github.com/sgl-project/sglang/issues/8833

Vicuna is a chat assistant trained by fine-tuning Llama 2 on user-shared conversations collected from ShareGPT. - Developed by: LMSYS - Model type: An auto-regressive language model based on the transformer architecture - License: Llama 2 Community License Agreement - Finetuned from model: Llama 2 - 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/ 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. - 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 Vicuna v1.5 is fine-tuned from Llama 2 with supervised instruction fine-tuning. The training data is around 125K conversations collected from ShareGPT.com. See more details in the "Training Details of Vicuna Models" section in the appendix of this paper. Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this paper and leaderboard.

NOTE: New version available Please check out a newer version of the weights here. Vicuna is a chat assistant trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT. - Developed by: LMSYS - Model type: An auto-regressive language model based on the transformer architecture. - License: Non-commercial license - Finetuned from model: LLaMA. - 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/ 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. - 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. Vicuna v1.3 is fine-tuned from LLaMA with supervised instruction fine-tuning. The training data is around 125K conversations collected from ShareGPT.com. See more details in the "Training Details of Vicuna Models" section in the appendix of this paper. Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this paper and leaderboard. Difference between different versions of Vicuna See vicunaweightsversion.md

This model is copied from https://huggingface.co/yuhuili/EAGLE-LLaMA3-Instruct-8B with the following modifications to make it compatible with SGLang: - Modify the architecture in config.json to LlamaForCausalLMEagle

NOTE: New version available Please check out a newer version of the weights here. Vicuna is a chat assistant trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT. - Developed by: LMSYS - Model type: An auto-regressive language model based on the transformer architecture. - License: Non-commercial license - Finetuned from model: LLaMA. - 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/ 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. - 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. Vicuna v1.3 is fine-tuned from LLaMA with supervised instruction fine-tuning. The training data is around 125K conversations collected from ShareGPT.com. See more details in the "Training Details of Vicuna Models" section in the appendix of this paper. Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this paper and leaderboard. Difference between different versions of Vicuna See vicunaweightsversion.md

Vicuna is a chat assistant trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT. - Developed by: LMSYS - Model type: An auto-regressive language model based on the transformer architecture. - License: Non-commercial license - Finetuned from model: LLaMA. - 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/ 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. - 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. Vicuna v1.3 is fine-tuned from LLaMA with supervised instruction fine-tuning. The training data is around 125K conversations collected from ShareGPT.com. See more details in the "Training Details of Vicuna Models" section in the appendix of this paper. Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this paper and leaderboard. Difference between different versions of Vicuna See vicunaweightsversion.md

NOTE: New version available Please check out a newer version of the weights here. NOTE: This "delta model" cannot be used directly. Users have to apply it on top of the original LLaMA weights to get actual Vicuna weights. See instructions. Vicuna is a chat assistant trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT. - Developed by: LMSYS - Model type: An auto-regressive language model based on the transformer architecture. - License: Non-commercial license - Finetuned from model: LLaMA. - 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/ 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. 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. Vicuna v1.1 is fine-tuned from LLaMA with supervised instruction fine-tuning. The training data is around 70K conversations collected from ShareGPT.com. See more details in the "Training Details of Vicuna Models" section in the appendix of this paper. Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this paper and leaderboard. Difference between different versions of Vicuna See vicunaweightsversion.md

Model Introduction The Eagle3 draft model was trained using the SpecForge framework for the Qwen3-235B-A22B model, leveraging a combination of UltraChat and ShareGPT datasets. Usage You can use this Eagle3 draft model in SGLang with the following command:

NOTE: New version available Please check out a newer version of the weights here. NOTE: This "delta model" cannot be used directly. Users have to apply it on top of the original LLaMA weights to get actual Vicuna weights. See instructions. Vicuna is a chat assistant trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT. - Developed by: LMSYS - Model type: An auto-regressive language model based on the transformer architecture. - License: Non-commercial license - Finetuned from model: LLaMA. - 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/ 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. 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. Vicuna v0 is fine-tuned from LLaMA with supervised instruction fine-tuning. The training data is around 70K conversations collected from ShareGPT.com. See more details in the "Training Details of Vicuna Models" section in the appendix of this paper. Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this paper and leaderboard. Difference between different versions of Vicuna See vicunaweightsversion.md

This model is copied from https://huggingface.co/yuhuili/EAGLE-LLaMA3-Instruct-70B with the following modifications to make it compatible with SGLang: - Modify the architecture in config.json to LlamaForCausalLMEagle

This model is copied from https://huggingface.co/yuhuili/EAGLE-llama2-chat-70B with the following modifications to make it compatible with SGLang: - Modify the architecture in config.json to LlamaForCausalLMEagle