chore(model gallery): add internlm_oreal-32b (#4872)

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2025-02-20 15:52:28 +01:00 committed by GitHub
parent ac4991b069
commit c27ce6c54d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -3959,6 +3959,22 @@
- filename: open-r1_OpenR1-Qwen-7B-Q4_K_M.gguf - filename: open-r1_OpenR1-Qwen-7B-Q4_K_M.gguf
sha256: d3bf99666cd19b637948ec9943044b591d3b906d0ee4f3ef1b3eb693ac8f66a6 sha256: d3bf99666cd19b637948ec9943044b591d3b906d0ee4f3ef1b3eb693ac8f66a6
uri: huggingface://bartowski/open-r1_OpenR1-Qwen-7B-GGUF/open-r1_OpenR1-Qwen-7B-Q4_K_M.gguf uri: huggingface://bartowski/open-r1_OpenR1-Qwen-7B-GGUF/open-r1_OpenR1-Qwen-7B-Q4_K_M.gguf
- !!merge <<: *qwen25
name: "internlm_oreal-32b"
urls:
- https://huggingface.co/internlm/OREAL-32B
- https://huggingface.co/bartowski/internlm_OREAL-32B-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-32B-Q4_K_M.gguf
files:
- filename: internlm_OREAL-32B-Q4_K_M.gguf
sha256: 5af1b3f66e3a1f95931a54500d03368c0cc7ca42cc67370338b29c18362e4a94
uri: huggingface://bartowski/internlm_OREAL-32B-GGUF/internlm_OREAL-32B-Q4_K_M.gguf
- &llama31 - &llama31
url: "github:mudler/LocalAI/gallery/llama3.1-instruct.yaml@master" ## LLama3.1 url: "github:mudler/LocalAI/gallery/llama3.1-instruct.yaml@master" ## LLama3.1
icon: https://avatars.githubusercontent.com/u/153379578 icon: https://avatars.githubusercontent.com/u/153379578