DeepSeek-R1-0528-Qwen3-8B

Home/Models/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B

DeepSeek-R1-0528-Qwen3-8B

deepseek-ai/DeepSeek-R1-0528-Qwen3-8B

kind: original
quant: bf16 (16-bit)
family: qwen
modality: text-generation
root model: deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
provider: deepseek-ai
license: mit
language: —
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 2,001,886
likes: 1,081
created: —
last modified: 2025-05-29T13:13:34+00:00 (2025)
huggingface: https://huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
tags: transformerssafetensorsqwen3text-generationconversationalarxiv:2501.12948license:mittext-generation-inferenceendpoints_compatibleregion:us

Benchmark scores (2)

benchmark	category	score	context	conf	evidence
GPQA Diamond	reasoning	62.0		high	`\| Qwen3-8B \| 76.0 \| 67.3 \| - \| 62.0 \| - \|`
LiveCodeBench	coding	60.5		high	`\| DeepSeek-R1-0528-Qwen3-8B \| 86.0 \| 76.3 \| 61.5 \| 61.1 \| 60.5 \|`

Discussions (27)

title	author	status	type
[BUG]: Garbled/Corrupted output when serving deepseek-ai/DeepSeek-R1-0528-Qwen3-8B with vllm/vllm-openai:v0.22.1-ubuntu2404	tianliang123456789	closed	disc
Update README.md	cherry0328	open	disc
Upload 8cfde66173265d81d161ff2b734120f0.jpg	Khaled0075	closed	disc
该模型Q4_K_M量化，在ollama部署该模型后不知所云	laaok	open	disc
This model was distilled using only SFT, or through a combination of SFT and RL?	wizardII	open	disc
Which Livecodebench version? v5? v6?	owao	open	disc
During git clone: encounters 2 file(s) that may not have been copied correctly on Windows	guynich	open	disc
Is this model can not use function calling？	Fatzard	open	disc
test_for_ds_r1_qwq_8b	JunZhangf	draft	disc
Failed to reproduce evaluation result on AIME24	cppowboy	closed	disc
A quick test comparing R1-0528-Qwen3-8B with Phi-4	gptlocalhost	open	disc
PR:DeepSeek-R1-0528	G-Nexus	closed	disc
ciudades turisticas	lolisponce	open	disc
Model collapse after SFT	Banjiuyufen	open	disc
Vocab missing tool-related strings in chat template, poor performance with tools	mattjcly	open	disc
Can you please release how you post-train qwen3 on deepseek?	ZeroWw	open	disc
Tried it, but not good as expected.	kk3dmax	open	disc
/no_think 标签不能用了吗	loong	open	disc
Any plans for a Qwen3-32B model?	wanghf	open	disc
BTW For programmer, `Gemma` series are best to help you write comments, docstrings, and documents.	DOFOFFICIAL	open	disc
QWEN3 is the bottleneck. QWEN3 can hardly understand prompts and commands and always `do something it likes`	DOFOFFICIAL	closed	disc
DeepSeek-R1-Lite	Dampfinchen	open	disc
generation_config.json is missing	Doctor-Chad-PhD	open	disc
Model broken	sm54	open	disc
牛啊牛啊	mrli008	open	disc
Any plans on gemma series? ;-;	Nakdesu	open	disc
Any plans on 30B-A3B model?	xxx777xxxASD	open	disc