github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-22 05:31:43 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #11244] Support for Baidu's ERNIE 4.5/VL #85092

New Issue
Open
opened 2026-05-09 22:31:53 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-05-09 22:31:53 -05:00
Owner
Copy Link

Originally created by @3unnycheung on GitHub (Jun 30, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11244

https://huggingface.co/collections/baidu/ernie-45-6861cd4c9be84540645f35c9

Image

Image

Originally created by @3unnycheung on GitHub (Jun 30, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11244 https://huggingface.co/collections/baidu/ernie-45-6861cd4c9be84540645f35c9 ![Image](https://github.com/user-attachments/assets/08d68683-0fe7-42db-bcfd-e060f9465e3a) ![Image](https://github.com/user-attachments/assets/d0425f4c-19a8-46c5-bb1e-2a9f62cad5ea)
GiteaMirror added the model label 2026-05-09 22:31:53 -05:00
GiteaMirror commented 2026-05-09 22:31:54 -05:00
Author
Owner
Copy Link

@dengcao commented on GitHub (Jul 3, 2025):

Error: unable to load model:https://huggingface.co/dengcao/ERNIE-4.5-0.3B-PT-GGUF

<!-- gh-comment-id:3032440572 --> @dengcao commented on GitHub (Jul 3, 2025): Error: unable to load model:https://huggingface.co/dengcao/ERNIE-4.5-0.3B-PT-GGUF
GiteaMirror commented 2026-05-09 22:31:55 -05:00
Author
Owner
Copy Link

@Himanshu8881212 commented on GitHub (Jul 4, 2025):

i read the paper where they mentioned that they have implemented a 4 bit and 2 bit lossless quantization. if we can get those models in ollama it would be something of a start. both the LLM and VLM

<!-- gh-comment-id:3036942034 --> @Himanshu8881212 commented on GitHub (Jul 4, 2025): i read the paper where they mentioned that they have implemented a 4 bit and 2 bit lossless quantization. if we can get those models in ollama it would be something of a start. both the LLM and VLM
GiteaMirror commented 2026-05-09 22:31:55 -05:00
Author
Owner
Copy Link

@Mugane commented on GitHub (Jul 6, 2025):

I prefer the reference model. We can quantize on our own

<!-- gh-comment-id:3040972831 --> @Mugane commented on GitHub (Jul 6, 2025): I prefer the reference model. We can quantize on our own
GiteaMirror commented 2026-05-09 22:31:56 -05:00
Author
Owner
Copy Link

@fighter3005 commented on GitHub (Jul 10, 2025):

Also the VL model would be very strong, as it sets Ernie apart from deepseek, qwen, magistral and the other reasoning models that don't support vision input, and puts it closer to the SOTA models like gemini 2.5 flash and gpt-4.1 or o1.

<!-- gh-comment-id:3057002342 --> @fighter3005 commented on GitHub (Jul 10, 2025): Also the VL model would be very strong, as it sets Ernie apart from deepseek, qwen, magistral and the other reasoning models that don't support vision input, and puts it closer to the SOTA models like gemini 2.5 flash and gpt-4.1 or o1.
GiteaMirror commented 2026-05-09 22:31:57 -05:00
Author
Owner
Copy Link

@fighter3005 commented on GitHub (Jul 18, 2025):

Quick question, llama.cpp will support the non VL versions soon. Possibly also the VL version at some point.
should we wait for that and use these implementations or should we implement model support for Ernie with the "new" model engine ourselves?

<!-- gh-comment-id:3088757572 --> @fighter3005 commented on GitHub (Jul 18, 2025): Quick question, llama.cpp will support the non VL versions soon. Possibly also the VL version at some point. should we wait for that and use these implementations or should we implement model support for Ernie with the "new" model engine ourselves?
GiteaMirror commented 2026-05-09 22:31:58 -05:00
Author
Owner
Copy Link

@3unnycheung commented on GitHub (Jul 19, 2025):

Quick question, llama.cpp will support the non VL versions soon. Possibly also the VL version at some point. should we wait for that and use these implementations or should we implement model support for Ernie with the "new" model engine ourselves?

https://github.com/ggml-org/llama.cpp/releases/tag/b5924

<!-- gh-comment-id:3092159328 --> @3unnycheung commented on GitHub (Jul 19, 2025): > Quick question, llama.cpp will support the non VL versions soon. Possibly also the VL version at some point. should we wait for that and use these implementations or should we implement model support for Ernie with the "new" model engine ourselves? https://github.com/ggml-org/llama.cpp/releases/tag/b5924
GiteaMirror commented 2026-05-09 22:31:59 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Aug 16, 2025):

$ ollama -v
ollama version is 0.11.5-rc2
$ ollama run ernie-4.5:21b-a3b-q4_K_M hello
Hello! 😊 How can I assist you today? Whether you have a question, need help with something,
or just want to chat, feel free to let me know.
<!-- gh-comment-id:3193044355 --> @rick-github commented on GitHub (Aug 16, 2025): ```console $ ollama -v ollama version is 0.11.5-rc2 $ ollama run ernie-4.5:21b-a3b-q4_K_M hello Hello! 😊 How can I assist you today? Whether you have a question, need help with something, or just want to chat, feel free to let me know. ```
GiteaMirror commented 2026-05-09 22:31:59 -05:00
Author
Owner
Copy Link

@fighter3005 commented on GitHub (Aug 16, 2025):

$ ollama -v
ollama version is 0.11.5-rc2
$ ollama run ernie-4.5:21b-a3b-q4_K_M hello
Hello! 😊 How can I assist you today? Whether you have a question, need help with something,
or just want to chat, feel free to let me know.

Is that the llama.cpp backend? Sadly they don't have vision support yet. :(

<!-- gh-comment-id:3193457317 --> @fighter3005 commented on GitHub (Aug 16, 2025): > $ ollama -v > ollama version is 0.11.5-rc2 > $ ollama run ernie-4.5:21b-a3b-q4_K_M hello > Hello! 😊 How can I assist you today? Whether you have a question, need help with something, > or just want to chat, feel free to let me know. Is that the llama.cpp backend? Sadly they don't have vision support yet. :(
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#85092
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?