model : register t_layer_inp for qwen3next - #25141
Merged
Merged
Conversation
Fix DFLASH for qwen-coder-next
This comment was marked as resolved.
This comment was marked as resolved.
Contributor
Author
Contributor
Author
|
Commit message was AI generated, code was manually checked & tested. DFlash was broken before and works now. See gist above. |
ruixiang63
reviewed
Jun 29, 2026
Added tensor for attention normalization in Qwen3 model.
ruixiang63
approved these changes
Jun 29, 2026
Member
Update OP to use our PR template please. |
jschmied
changed the title
Contributor
Author
|
local test run is green, dont know how to fix failing checks. Please help. |
Member
Ignore them, they are unrelated, but please update your OP so that is using our PR template. |
Contributor
Author
|
changed PR description, hope its better now |
Member
Please check how other PRs are formatted. You are missing "Requirements". |
Contributor
Author
|
added, sorry |
CISC
approved these changes
Jun 30, 2026
1 task
turbo-tan
pushed a commit
to turbo-tan/llama.cpp-tq3
that referenced
this pull request
Jul 1, 2026
* Fix input assignment in layer processing loop Fix DFLASH for qwen-coder-next * add line break Added tensor for attention normalization in Qwen3 model.
DFveloper
pushed a commit
to DFveloper/aikar-engine
that referenced
this pull request
Jul 3, 2026
* Fix input assignment in layer processing loop Fix DFLASH for qwen-coder-next * add line break Added tensor for attention normalization in Qwen3 model.
LyndonBlack
added a commit
to LyndonBlack/llama.cpp-Ternary-1.58Bit-and-TurboQuant
that referenced
this pull request
Jul 5, 2026
Sync with upstream (44 commits): - DFlash speculative decoding support (ggml-org#22105, ggml-org#25110, ggml-org#25246) - DeepSeek V4 model support (ggml-org#24162) - MoE topk fusion for 288 experts (ggml-org#25267) - Qwen3Next model fixes (ggml-org#25141) - CUDA FA optimizations + Gemma E4B MTP fix - NVFP4 support for WebGPU, AVX2 nvfp4 dot - llama_model_ftype_name() API - Various UI, server, and build fixes Resolved conflicts: - arch-fallback.h: kept Q2_0 vec_dot generic defines - llama-kv-cache-iswa.{h,cpp}: merged upstream 2-ctor refactor with turboquant layer_type_k_cb - llama-model-loader.cpp: adopted upstream llama_ftype_name() with Q2_0 entry preserved Assisted-by: Hermes Agent
adrianhoehne
pushed a commit
to adrianhoehne/llama.cpp
that referenced
this pull request
Jul 5, 2026
* Fix input assignment in layer processing loop Fix DFLASH for qwen-coder-next * add line break Added tensor for attention normalization in Qwen3 model.
zommiommy
pushed a commit
to zommiommy/llama.cpp
that referenced
this pull request
Aug 18, 2026
* Fix input assignment in layer processing loop Fix DFLASH for qwen-coder-next * add line break Added tensor for attention normalization in Qwen3 model.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Overview
This fixes speculative decoding support for
qwen3next.qwen3nextgraphs did not setres->t_layer_inp[il], so drafters that extract target hidden states, such as DFlash and EAGLE3, abort at runtime:Requirements