[GH-ISSUE #6742] Add OLMoE 1b-7b #50760

Closed
opened 2026-04-28 17:01:57 -05:00 by GiteaMirror · 4 comments
Owner

Originally created by @Meshwa428 on GitHub (Sep 11, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6742

It's the best for mobile applications and can run on edge devices with only 1b active params

Hf reference: https://huggingface.co/allenai/OLMoE-1B-7B-0924-Instruct

Motivation:
It is fast, reduces carbon emissions and runs on edge devices. What else do we need 😋?

With only 1b active params it is on par with other llms

Originally created by @Meshwa428 on GitHub (Sep 11, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6742 It's the best for mobile applications and can run on edge devices with only 1b active params Hf reference: https://huggingface.co/allenai/OLMoE-1B-7B-0924-Instruct Motivation: It is fast, reduces carbon emissions and runs on edge devices. What else do we need 😋? With only 1b active params it is on par with other llms
GiteaMirror added the model label 2026-04-28 17:01:57 -05:00
Author
Owner

@Meshwa428 commented on GitHub (Sep 13, 2024):

Update:

Gguf quants for this model can be found here:
Meshwa/OLMoE-1b-7b-0924-Instruct-gguf

These quants are created using this llama.cpp fork whose merge is pending
#9462

Repo: https://github.com/2015aroras/llama.cpp/tree/olmoe

<!-- gh-comment-id:2350068620 --> @Meshwa428 commented on GitHub (Sep 13, 2024): Update: Gguf quants for this model can be found here: [**Meshwa/OLMoE-1b-7b-0924-Instruct-gguf**](https://huggingface.co/Meshwa/OLMoE-1b-7b-0924-Instruct-gguf) These quants are created using this llama.cpp fork whose merge is pending [#9462](https://github.com/ggerganov/llama.cpp/pull/9462) Repo: https://github.com/2015aroras/llama.cpp/tree/olmoe
Author
Owner

@Meshwa428 commented on GitHub (Sep 14, 2024):

Any updates?

<!-- gh-comment-id:2351080437 --> @Meshwa428 commented on GitHub (Sep 14, 2024): Any updates?
Author
Owner

@Meshwa428 commented on GitHub (Sep 18, 2024):

OLMoE 1b-7b has been finally implemented in llama.cpp - master branch

Thanks🙏

<!-- gh-comment-id:2359043396 --> @Meshwa428 commented on GitHub (Sep 18, 2024): OLMoE 1b-7b has been finally implemented in [llama.cpp - master branch](https://github.com/ggerganov/llama.cpp) Thanks🙏
Author
Owner

@dhiltgen commented on GitHub (Sep 25, 2024):

Let's track this via #2337

<!-- gh-comment-id:2375352061 --> @dhiltgen commented on GitHub (Sep 25, 2024): Let's track this via #2337
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#50760