Merged
Conversation
ggerganov
requested changes
Mar 12, 2023
Closed
ggerganov
approved these changes
Mar 12, 2023
Hades32
pushed a commit
to Hades32/llama.cpp
that referenced
this pull request
Mar 21, 2023
CI fine tuning
theo77186
pushed a commit
to theo77186/llama.cpp
that referenced
this pull request
Oct 28, 2025
jesusmb1995
pushed a commit
to jesusmb1995/llama.cpp
that referenced
this pull request
Oct 30, 2025
implement logging redirection
SamuelOliveirads
pushed a commit
to SamuelOliveirads/llama.cpp
that referenced
this pull request
Dec 29, 2025
Ref ggml-org#29 Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
rururush
pushed a commit
to USTC-ADSL/llama.cpp
that referenced
this pull request
Mar 16, 2026
* print build type * wip * print compiling flags * wip * wip
TheTom
referenced
this pull request
in TheTom/llama-cpp-turboquant
Mar 25, 2026
…gml-org#31 Block 128: PPL=165.6 (same as block 32) Disabled Q rotation: PPL=165.6 (same) Root cause: dynamic_cast fails for MoE hybrid memory context. Q rotation and V inverse rotation never execute. Co-Authored-By: tturney@psyguard.ai Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
TheTom
referenced
this pull request
in TheTom/llama-cpp-turboquant
Mar 25, 2026
…#30 ROOT CAUSE: pre-rotate-queries never executed because: 1. Q ne[0]=256 (GQA concatenated heads), rotation matrix ne[0]=128 2. mctx dynamic_cast failed for MoE hybrid memory FIX: put inverse WHT rotation back in dequantize_full_block. This is slower (10.7 tok/s vs 77.7) but produces CORRECT results. PERPLEXITY RESULTS: - f16: 6.121 - q8_0: 6.111 - q4_0: 6.142 - turbo3: 6.194 (+1.2% vs q8_0) ✅ The speed optimization (pre-rotate-queries) needs to be reimplemented to work with GQA head layout and hybrid memory types. Co-Authored-By: tturney@psyguard.ai Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
TheTom
referenced
this pull request
in TheTom/llama-cpp-turboquant
Mar 26, 2026
…31 Block 128: PPL=165.6 (same as block 32) Disabled Q rotation: PPL=165.6 (same) Root cause: dynamic_cast fails for MoE hybrid memory context. Q rotation and V inverse rotation never execute. Co-Authored-By: tturney@psyguard.ai Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
TheTom
referenced
this pull request
in TheTom/llama-cpp-turboquant
Mar 26, 2026
…#30 ROOT CAUSE: pre-rotate-queries never executed because: 1. Q ne[0]=256 (GQA concatenated heads), rotation matrix ne[0]=128 2. mctx dynamic_cast failed for MoE hybrid memory FIX: put inverse WHT rotation back in dequantize_full_block. This is slower (10.7 tok/s vs 77.7) but produces CORRECT results. PERPLEXITY RESULTS: - f16: 6.121 - q8_0: 6.111 - q4_0: 6.142 - turbo3: 6.194 (+1.2% vs q8_0) ✅ The speed optimization (pre-rotate-queries) needs to be reimplemented to work with GQA head layout and hybrid memory types. Co-Authored-By: tturney@psyguard.ai Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
spiritbuun
referenced
this pull request
in spiritbuun/llama-cpp-turboquant-cuda
Mar 27, 2026
- turbo4 K+V results on Qwen3.5-27B (-0.32% vs q8_0) and Qwen3-14B (+6.3%) - Sparse V dequant benchmarks: MoE native dequant +10.9% at 8K - Gemma-3 turbo3 results post-iSWA fix (+3.3%) - KVLinC no-K-rotation negative result - Speculative decoding negative result - CUDA 13.2 compatibility verified - Experiments TheTom#31, TheTom#39, TheTom#42, TheTom#45, ggml-org#49, ggml-org#50, ggml-org#51 status updates Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
didlawowo
pushed a commit
to didlawowo/llama.cpp
that referenced
this pull request
Mar 27, 2026
…ml-org#31 ggml-org#30 ROOT CAUSE: pre-rotate-queries never executed because: 1. Q ne[0]=256 (GQA concatenated heads), rotation matrix ne[0]=128 2. mctx dynamic_cast failed for MoE hybrid memory FIX: put inverse WHT rotation back in dequantize_full_block. This is slower (10.7 tok/s vs 77.7) but produces CORRECT results. PERPLEXITY RESULTS: - f16: 6.121 - q8_0: 6.111 - q4_0: 6.142 - turbo3: 6.194 (+1.2% vs q8_0) ✅ The speed optimization (pre-rotate-queries) needs to be reimplemented to work with GQA head layout and hybrid memory types. Co-Authored-By: tturney@psyguard.ai Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
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
2 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.
This would be the initial PR to be able to compile stuff in Windows.
In particular, MSVC is very picky about the features you can use and you cannot.
With C++11
A PR for the CMake build system (as agreed in #22) will be separated.
These changes were tested with MSVC 19.34.31937.0 (VS Studio 2022) and in macOS 12.6 with
Apple clang version 13.1.6