diff --git a/gallery/index.yaml b/gallery/index.yaml index ed401325..1d4f98fc 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -6266,6 +6266,22 @@ - filename: agentica-org_DeepScaleR-1.5B-Preview-Q4_K_M.gguf sha256: bf51b412360a84792ae9145e2ca322379234c118dbff498ff08e589253b67ded uri: huggingface://bartowski/agentica-org_DeepScaleR-1.5B-Preview-GGUF/agentica-org_DeepScaleR-1.5B-Preview-Q4_K_M.gguf +- !!merge <<: *deepseek-r1 + name: "internlm_oreal-deepseek-r1-distill-qwen-7b" + urls: + - https://huggingface.co/internlm/OREAL-DeepSeek-R1-Distill-Qwen-7B + - https://huggingface.co/bartowski/internlm_OREAL-DeepSeek-R1-Distill-Qwen-7B-GGUF + description: | + We introduce OREAL-7B and OREAL-32B, a mathematical reasoning model series trained using Outcome REwArd-based reinforcement Learning, a novel RL framework designed for tasks where only binary outcome rewards are available. + + With OREAL, a 7B model achieves 94.0 pass@1 accuracy on MATH-500, matching the performance of previous 32B models. OREAL-32B further surpasses previous distillation-trained 32B models, reaching 95.0 pass@1 accuracy on MATH-500. + overrides: + parameters: + model: internlm_OREAL-DeepSeek-R1-Distill-Qwen-7B-Q4_K_M.gguf + files: + - filename: internlm_OREAL-DeepSeek-R1-Distill-Qwen-7B-Q4_K_M.gguf + sha256: fa9dc8b0d4be0952252c25ff33e766a8399ce7b085647b95abe3edbe536cd8ed + uri: huggingface://bartowski/internlm_OREAL-DeepSeek-R1-Distill-Qwen-7B-GGUF/internlm_OREAL-DeepSeek-R1-Distill-Qwen-7B-Q4_K_M.gguf - &qwen2 url: "github:mudler/LocalAI/gallery/chatml.yaml@master" ## Start QWEN2 name: "qwen2-7b-instruct"