Skip to content

Commit 04764bb

Browse files
localai-botmudler
andauthored
chore(model gallery): 🤖 add 1 new models via gallery agent (#11268)
chore(model gallery): 🤖 add new models via gallery agent Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com>
1 parent 2f3dd40 commit 04764bb

1 file changed

Lines changed: 44 additions & 0 deletions

File tree

gallery/index.yaml

Lines changed: 44 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,48 @@
11
---
2+
- name: "deepseek-v4-flash-0731"
3+
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
4+
urls:
5+
- https://huggingface.co/unsloth/DeepSeek-V4-Flash-0731-GGUF
6+
description: |
7+
# DeepSeek-V4-Flash-0731
8+
9+
Technical Report👁️
10+
11+
## Introduction
12+
13+
**DeepSeek-V4-Flash-0731** is the official release of **DeepSeek-V4-Flash**, superseding the preview version, with substantially enhanced agentic capabilities. It has the same model structure as DeepSeek-V4-Flash-DSpark, i.e. it comes with a speculative decoding module attached.
14+
15+
DeepSeek-V4-Flash-0731 outperforms DeepSeek-V4-Pro (Preview) on benchmarks listed below despite its far smaller activated parameter count, and is broadly competitive with the strongest proprietary models available.
16+
17+
Notes:
18+
19+
1. For the Code Agent tasks among the public benchmarks above, DeepSeek-V4-Flash-0731 is evaluated with the minimal mode of DeepSeek Harness (to be released) as the agent framework, using the `max` reasoning effort level with `temperature = 1.0, top_p = 0.95`.
20+
2. † DSBench-FullStack is an internal full-stack development test set; DSBench-Hard is an internal test set of difficult coding-agent problems.
21+
22+
## Chat Template
23+
24+
...
25+
license: "mit"
26+
tags:
27+
- llm
28+
- gguf
29+
- deepseek
30+
icon: https://github.com/deepseek-ai/DeepSeek-V2/blob/main/figures/logo.svg
31+
overrides:
32+
backend: ds4
33+
function:
34+
grammar:
35+
disable: true
36+
known_usecases:
37+
- chat
38+
parameters:
39+
model: ds4flash.gguf
40+
template:
41+
use_tokenizer_template: true
42+
files:
43+
- filename: ds4flash.gguf
44+
sha256: ""
45+
uri: https://huggingface.co/unsloth/DeepSeek-V4-Flash-0731-GGUF
246
- name: "parable-granite-4.1-3b-claude-fable-5"
347
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
448
urls:

0 commit comments

Comments
 (0)