github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-22 13:42:25 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #2802] Madlad400 model #1695

New Issue
Open
opened 2026-04-12 11:40:08 -05:00 by GiteaMirror · 7 comments
GiteaMirror commented 2026-04-12 11:40:08 -05:00
Owner
Copy Link

Originally created by @malipetek on GitHub (Feb 28, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/2802

Hello, I wanted to test madlad400 which said to be a great translator model.

I downloaded the GGUF and created a file with models name with only FROM line. It looks like model created but when I test-run it, it outputs 2 empty lines for some reason. And when I specify --verbose flag there is no report at the end of inference.

This should work in theory no?
image

Originally created by @malipetek on GitHub (Feb 28, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/2802 Hello, I wanted to test [madlad400](https://huggingface.co/jbochi/madlad400-3b-mt/blob/main/model-q4k.gguf) which said to be a great translator model. I downloaded the GGUF and created a file with models name with only FROM line. It looks like model created but when I test-run it, it outputs 2 empty lines for some reason. And when I specify `--verbose` flag there is no report at the end of inference. This should work in theory no? <img width="842" alt="image" src="https://github.com/ollama/ollama/assets/13527277/f1cc65ea-4dd1-4c58-adc4-eb198f087dd3">
GiteaMirror added the model label 2026-04-12 11:40:08 -05:00
GiteaMirror commented 2026-04-12 11:40:10 -05:00
Author
Owner
Copy Link

@easp commented on GitHub (Feb 29, 2024):

The model architecture isn't supported, so it won't work.

You can try adding your voice to this issue on the llama.cpp repo (ollama uses llama.cpp)
https://github.com/ggerganov/llama.cpp/issues/4316

<!-- gh-comment-id:1970306210 --> @easp commented on GitHub (Feb 29, 2024): The model architecture isn't supported, so it won't work. You can try adding your voice to this issue on the llama.cpp repo (ollama uses llama.cpp) https://github.com/ggerganov/llama.cpp/issues/4316
GiteaMirror commented 2026-04-12 11:40:10 -05:00
Author
Owner
Copy Link

@flostellbrink commented on GitHub (Jul 6, 2024):

Llama.cpp just added support for T5 architectures: https://github.com/ggerganov/llama.cpp/pull/8141
Is this possible now?

<!-- gh-comment-id:2211758776 --> @flostellbrink commented on GitHub (Jul 6, 2024): Llama.cpp just added support for T5 architectures: https://github.com/ggerganov/llama.cpp/pull/8141 Is this possible now?
GiteaMirror commented 2026-04-12 11:40:11 -05:00
Author
Owner
Copy Link

@malipetek commented on GitHub (Jul 7, 2024):

I tried the same thing with ollama version 0.1.48 with no luck, how can we tell which ollama version is using which Llama.cpp?
image

<!-- gh-comment-id:2212510891 --> @malipetek commented on GitHub (Jul 7, 2024): I tried the same thing with ollama version `0.1.48` with no luck, how can we tell which ollama version is using which Llama.cpp? <img width="842" alt="image" src="https://github.com/ollama/ollama/assets/13527277/26b9d11c-f16b-431c-b90f-82097a64d2c3">
GiteaMirror commented 2026-04-12 11:40:12 -05:00
Author
Owner
Copy Link

@malipetek commented on GitHub (Jul 11, 2024):

This still does not work with v2.0.1, can someone knowledgeable clarify why? Ollama has Llama.cpp as a submodule, so when builds are made it should be having the latest version of Llama.cpp right?

<!-- gh-comment-id:2224041470 --> @malipetek commented on GitHub (Jul 11, 2024): This still does not work with `v2.0.1`, can someone knowledgeable clarify why? Ollama has `Llama.cpp` as a submodule, so when builds are made it should be having the latest version of `Llama.cpp` right?
GiteaMirror commented 2026-04-12 11:40:12 -05:00
Author
Owner
Copy Link

@tritiumoxide commented on GitHub (Jul 13, 2024):

@malipetek The problem is that ollama uses llama-server and even in the latest version of llama.cpp support for T5 was so far only added to llama-cli. Support for T5 in llama-server is still a TODO. Even once it's added, there might be some API changes that ollama will have to reflect, introducing some more lag, especially since T5 is an encoder-decoder model, and not decoder-only like most previously supported models.

<!-- gh-comment-id:2227096278 --> @tritiumoxide commented on GitHub (Jul 13, 2024): @malipetek The problem is that ollama uses llama-server and even in the latest version of llama.cpp support for T5 was so far only added to llama-cli. Support for T5 in llama-server is still a TODO. Even once it's added, there might be some API changes that ollama will have to reflect, introducing some more lag, especially since T5 is an encoder-decoder model, and not decoder-only like most previously supported models.
GiteaMirror commented 2026-04-12 11:40:12 -05:00
Author
Owner
Copy Link

@malipetek commented on GitHub (Jul 15, 2024):

@tritiumoxide Thanks for the explanation.

<!-- gh-comment-id:2229173098 --> @malipetek commented on GitHub (Jul 15, 2024): @tritiumoxide Thanks for the explanation.
GiteaMirror commented 2026-04-12 11:40:13 -05:00
Author
Owner
Copy Link

@aymbot commented on GitHub (Sep 16, 2024):

any updates regarding this model request?

<!-- gh-comment-id:2352689029 --> @aymbot commented on GitHub (Sep 16, 2024): any updates regarding this model request?
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#1695
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?