github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #4245] DeepSeek-V2 #2647

New Issue
Closed
opened 2026-04-12 12:59:30 -05:00 by GiteaMirror · 6 comments
GiteaMirror commented 2026-04-12 12:59:30 -05:00
Owner
Copy Link

Originally created by @enryteam on GitHub (May 8, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4245

https://huggingface.co/deepseek-ai/DeepSeek-V2

thanks

Originally created by @enryteam on GitHub (May 8, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4245 https://huggingface.co/deepseek-ai/DeepSeek-V2 thanks
GiteaMirror added the model label 2026-04-12 12:59:30 -05:00
GiteaMirror commented 2026-04-12 12:59:31 -05:00
Author
Owner
Copy Link

@taozhiyuai commented on GitHub (May 8, 2024):

deepseek team does not offer gguf, which in not in their plan at the moment. and its performance is not much better than llama-3-70b, but 3 times bigger in parameters than llama 3 70b. I suggest personally abort it. and it is over 200B parameters which is very hard to be loaded in most GPUs.

<!-- gh-comment-id:2099811975 --> @taozhiyuai commented on GitHub (May 8, 2024): deepseek team does not offer gguf, which in not in their plan at the moment. and its performance is not much better than llama-3-70b, but 3 times bigger in parameters than llama 3 70b. I suggest personally abort it. and it is over 200B parameters which is very hard to be loaded in most GPUs.
GiteaMirror commented 2026-04-12 12:59:31 -05:00
Author
Owner
Copy Link

@mattapperson commented on GitHub (May 9, 2024):

@taozhiyuai I think you are mistaken. This model is MoE to yes 200b params but only ~20B are active at a time making inference much faster

<!-- gh-comment-id:2102557112 --> @mattapperson commented on GitHub (May 9, 2024): @taozhiyuai I think you are mistaken. This model is MoE to yes 200b params but only ~20B are active at a time making inference much faster
GiteaMirror commented 2026-04-12 12:59:32 -05:00
Author
Owner
Copy Link

@tiwentichat commented on GitHub (May 17, 2024):

deepseek team does not offer gguf, which in not in their plan at the moment. and its performance is not much better than llama-3-70b, but 3 times bigger in parameters than llama 3 70b. I suggest personally abort it. and it is over 200B parameters which is very hard to be loaded in most GPUs.

Deepseek v2 chat is better than Llama 3 70B in Chinese question.

Q list:

三苏是谁

长河落日圆上一句是什么

长风几万里下一句是什么 （70b failed 1 time, correct another time,

滕王阁序是杜甫写的吗？ (70b say no, then tell me the author is 欧阳修..........

and 70b's answer is mixing EN and Chinese together, deepseek v2 chat, i didn't see that by now.

<!-- gh-comment-id:2116448347 --> @tiwentichat commented on GitHub (May 17, 2024): > deepseek team does not offer gguf, which in not in their plan at the moment. and its performance is not much better than llama-3-70b, but 3 times bigger in parameters than llama 3 70b. I suggest personally abort it. and it is over 200B parameters which is very hard to be loaded in most GPUs. Deepseek v2 chat is better than Llama 3 70B in Chinese question. Q list: 三苏是谁 长河落日圆上一句是什么 长风几万里下一句是什么 （70b failed 1 time, correct another time, 滕王阁序是杜甫写的吗？ (70b say no, then tell me the author is 欧阳修.......... and 70b's answer is mixing EN and Chinese together, deepseek v2 chat, i didn't see that by now.
GiteaMirror commented 2026-04-12 12:59:32 -05:00
Author
Owner
Copy Link

@KongNan commented on GitHub (May 25, 2024):

deepseek team does not offer gguf

Will the GGUF offered by others work?
https://huggingface.co/leafspark/DeepSeek-V2-Chat-GGUF

<!-- gh-comment-id:2130547937 --> @KongNan commented on GitHub (May 25, 2024): > deepseek team does not offer gguf Will the GGUF offered by others work? https://huggingface.co/leafspark/DeepSeek-V2-Chat-GGUF
GiteaMirror commented 2026-04-12 12:59:32 -05:00
Author
Owner
Copy Link

@100ZZ commented on GitHub (May 28, 2024):

deepseek team does not offer gguf

Will the GGUF offered by others work? https://huggingface.co/leafspark/DeepSeek-V2-Chat-GGUF

我试过下载别人转换的GGUF都没发运行；然后huggingface下载原始的，通过llama.cpp，貌似arch不支持

<!-- gh-comment-id:2135318326 --> @100ZZ commented on GitHub (May 28, 2024): > > deepseek team does not offer gguf > > Will the GGUF offered by others work? https://huggingface.co/leafspark/DeepSeek-V2-Chat-GGUF 我试过下载别人转换的GGUF都没发运行；然后huggingface下载原始的，通过llama.cpp，貌似arch不支持
GiteaMirror commented 2026-04-12 12:59:33 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Jun 1, 2024):

Hi there, this should here now: https://ollama.com/library/deepseek-v2

<!-- gh-comment-id:2143309172 --> @jmorganca commented on GitHub (Jun 1, 2024): Hi there, this should here now: https://ollama.com/library/deepseek-v2
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label model
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#2647
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?