mirror/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2026-07-10 08:18:33 -04:00
Code Issues Packages Projects Releases Wiki Activity

chore(model-gallery): ⬆️ update checksum (#10749)

Browse Source 
⬆️ Checksum updates in gallery/index.yaml

Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: mudler <2420543+mudler@users.noreply.github.com>
This commit is contained in:
LocalAI [bot]
2026-07-09 01:12:22 +02:00
committed by GitHub
parent 40dae953f4
commit e948f27965
1 changed files with 2 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
gallery/index.yaml
View File

@@ -3,22 +3,7 @@
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
urls:
- https://huggingface.co/unsloth/DeepSeek-V4-Flash-GGUF
description: |
# DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
Technical Report👁️
## Introduction
We present a preview version of **DeepSeek-V4** series, including two strong Mixture-of-Experts (MoE) language models — **DeepSeek-V4-Pro** with 1.6T parameters (49B activated) and **DeepSeek-V4-Flash** with 284B parameters (13B activated) — both supporting a context length of **one million tokens**.
DeepSeek-V4 series incorporate several key upgrades in architecture and optimization:
1. **Hybrid Attention Architecture:** We design a hybrid attention mechanism combining Compressed Sparse Attention (CSA) and Heavily Compressed Attention (HCA) to dramatically improve long-context efficiency. In the 1M-token context setting, DeepSeek-V4-Pro requires only **27% of single-token inference FLOPs** and **10% of KV cache** compared with DeepSeek-V3.2.
2. **Manifold-Constrained Hyper-Connections (mHC):** We incorporate mHC to strengthen conventional residual connections, enhancing stability of signal propagation across layers while preserving model expressivity.
3. **Muon Optimizer:** We employ the Muon optimizer for faster convergence and greater training stability.
...
description: "# DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence\n\nTechnical Report\U0001F441️\n\n## Introduction\n\nWe present a preview version of **DeepSeek-V4** series, including two strong Mixture-of-Experts (MoE) language models — **DeepSeek-V4-Pro** with 1.6T parameters (49B activated) and **DeepSeek-V4-Flash** with 284B parameters (13B activated) — both supporting a context length of **one million tokens**.\n\nDeepSeek-V4 series incorporate several key upgrades in architecture and optimization:\n\n1. **Hybrid Attention Architecture:** We design a hybrid attention mechanism combining Compressed Sparse Attention (CSA) and Heavily Compressed Attention (HCA) to dramatically improve long-context efficiency. In the 1M-token context setting, DeepSeek-V4-Pro requires only **27% of single-token inference FLOPs** and **10% of KV cache** compared with DeepSeek-V3.2.\n2. **Manifold-Constrained Hyper-Connections (mHC):** We incorporate mHC to strengthen conventional residual connections, enhancing stability of signal propagation across layers while preserving model expressivity.\n3. **Muon Optimizer:** We employ the Muon optimizer for faster convergence and greater training stability.\n\n...\n"
license: "mit"
tags:
- llm
@@ -38,8 +23,8 @@
use_tokenizer_template: true
files:
- filename: ds4flash.gguf
sha256: ""
uri: https://huggingface.co/unsloth/DeepSeek-V4-Flash-GGUF
sha256: 58a206328080d51b0a374b1e4684c173e3e151b9282be47ee935bfb27047ddfa
- name: "qwopus3.6-35b-a3b-coder-mtp"
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
urls: