Weighted-Reward Preference Optimization for Implicit Model Fusion
Paper
•
2412.03187
•
Published
•
12
FuseChat 3.0
Preference Optimization for Implicit Model Fusion
-
FuseAI/FuseChat-Llama-3.1-8B-Instruct
Text Generation • Updated • 220 • 10Note Final DPO version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Llama-3.1-8B-Instruct.
FuseAI/FuseChat-Llama-3.2-3B-Instruct
Updated • 88 • 6Note Final DPO version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Llama-3.2-3B-Instruct.
FuseAI/FuseChat-Llama-3.2-1B-Instruct
Updated • 45 • 4Note Final DPO version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Llama-3.2-1B-Instruct.
FuseAI/FuseChat-Qwen-2.5-7B-Instruct
Updated • 125 • 12Note Final DPO version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Qwen-2.5-7B-Instruct.
FuseAI/FuseChat-Gemma-2-9B-Instruct
Updated • 60 • 7Note Final DPO version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Gemma-2-9B-Instruct.
FuseAI/FuseChat-Llama-3.1-8B-SFT
Updated • 163 • 1Note SFT version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Llama-3.1-8B-Instruct.
FuseAI/FuseChat-Llama-3.2-3B-SFT
Updated • 120 • 3Note SFT version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Llama-3.2-3B-Instruct.
FuseAI/FuseChat-Llama-3.2-1B-SFT
Updated • 43Note SFT version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Llama-3.2-1B-Instruct.
FuseAI/FuseChat-Qwen-2.5-7B-SFT
Updated • 39 • 2Note SFT version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Qwen-2.5-7B-Instruct.
FuseAI/FuseChat-Gemma-2-9B-SFT
Updated • 59 • 4Note SFT version of FuseChat-3.0. Source LLMs: Gemma-2-27B-It, Mistral-Large-Instruct-2407, Qwen-2.5-72B-Instruct, and Llama-3.1-70B-Instruct. Target LLM: Gemma-2-9B-Instruct.
FuseAI/FuseChat-3.0-SFT-Data
Viewer • Updated • 94.5k • 179 • 1Note SFT dataset for FuseChat-3.0.
FuseAI/FuseChat-3.0-DPO-Data
Viewer • Updated • 64.1k • 178Note DPO dataset for FuseChat-3.0.
-
FuseChat-3.0: Preference Optimization Meets Heterogeneous Model Fusion
Paper • 2503.04222 • Published • 12