[GH-ISSUE #13275] Model request: Qwen3-Next #70833

Closed
opened 2026-05-04 23:08:37 -05:00 by GiteaMirror · 7 comments
Owner

Originally created by @yqchen-sci on GitHub (Nov 30, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13275

Unsloth has already released the qwen-next-80b model in GGUF format.

Originally created by @yqchen-sci on GitHub (Nov 30, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13275 Unsloth has already released the [qwen-next-80b model](https://huggingface.co/unsloth/Qwen3-Next-80B-A3B-Instruct-GGUF) in GGUF format.
GiteaMirror added the model label 2026-05-04 23:08:37 -05:00
Author
Owner

@rick-github commented on GitHub (Nov 30, 2025):

https://github.com/ggml-org/llama.cpp/pull/16095

The next vendor sync after b7186 will make support available in ollama.

Be aware that this is just initial model support, and that speed optimization still needs to be done.

<!-- gh-comment-id:3592540192 --> @rick-github commented on GitHub (Nov 30, 2025): https://github.com/ggml-org/llama.cpp/pull/16095 The next vendor sync after [b7186](https://github.com/ggml-org/llama.cpp/releases/tag/b7186) will make support available in ollama. Be aware that this is just initial model support, and that speed optimization still needs to be done.
Author
Owner

@Ubong000 commented on GitHub (Dec 3, 2025):

When do you except it to be vendor synced? :)

<!-- gh-comment-id:3606723923 --> @Ubong000 commented on GitHub (Dec 3, 2025): When do you except it to be vendor synced? :)
Author
Owner

@mehditahmasebi commented on GitHub (Dec 3, 2025):

++++++++

<!-- gh-comment-id:3607889285 --> @mehditahmasebi commented on GitHub (Dec 3, 2025): ++++++++
Author
Owner

@rick-github commented on GitHub (Dec 3, 2025):

https://github.com/ollama/ollama/pull/12992 is targeting b7209 so if that is merged, ollama will support qwen3-next.

<!-- gh-comment-id:3607913613 --> @rick-github commented on GitHub (Dec 3, 2025): https://github.com/ollama/ollama/pull/12992 is targeting [b7209](https://github.com/ggml-org/llama.cpp/releases/tag/b7209) so if that is merged, ollama will support qwen3-next.
Author
Owner

@changjiandenuoen commented on GitHub (Dec 4, 2025):

hope this feature out soon

<!-- gh-comment-id:3611103965 --> @changjiandenuoen commented on GitHub (Dec 4, 2025): hope this feature out soon
Author
Owner

@Ubong000 commented on GitHub (Dec 4, 2025):

Its working with the v0.13.2 rc0 version, but very unoptimizied. Got around 20 t/s Q4 with an RTX A6000

<!-- gh-comment-id:3611335519 --> @Ubong000 commented on GitHub (Dec 4, 2025): Its working with the v0.13.2 rc0 version, but very unoptimizied. Got around 20 t/s Q4 with an RTX A6000
Author
Owner

@rick-github commented on GitHub (Dec 9, 2025):

https://ollama.com/library/qwen3-next

<!-- gh-comment-id:3629709289 --> @rick-github commented on GitHub (Dec 9, 2025): https://ollama.com/library/qwen3-next
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#70833