github llama.cpp · just now · inference

llama.cpp b10676 fixes conv_transpose_2d for multiple batches in ggml and Metal

from English

The llama.cpp release b10676 corrects a bug in `conv_transpose_2d` where only the first batch was computed, leaving subsequent batches as zero. This fix applies to both the core ggml library and the Metal backend implementation.

ggml: The source permutation and compute loop now iterate over the batch dimension, and work buffer size is scaled by the batch count.
Metal: The kernel grid covers all batches, decoding the batch index from the z coordinate to offset input and destination indices.
A multi-batch test case was added to `test-backend-ops` to verify the fix.

This ensures that multi-batch convolution results are correct across CPU and GPU backends.

Importance 1/3 Trust 1/3 llama.cpp Hugging Face Inference efficiency

Read original

github llama.cpp · 2h ago · 1 view

llama.cpp b10673 adds fa-vec tunings for Apple M4

The llama.cpp project released build b10673, which includes specific performance optimizations for Apple M4 hardware. This update introduces fa_vec_tuned_table records into the ggml-metal-tuning module to improve Metal backend efficiency.

github llama.cpp · 3h ago · 2 views

llama.cpp b10670 routes quantized KV decode to TILE on Xe2 (BMG)

The llama.cpp project released version b10670, which includes a specific optimization for Intel's Xe2 architecture. The update routes the quantized key-value (KV) decode operation to use the TILE backend exclusively for BMG hardware, while keeping the VEC backend on other architectures until they are validated.

github llama.cpp · 4h ago · 2 views

llama.cpp b10673 adds fa-vec tunings for Apple M4

llama.cpp b10670 routes quantized KV decode to TILE on Xe2 (BMG)

llama.cpp b10672 adds Whisper.cpp support, Qwen3.5 NPU enablement, and OpenVINO 2026.3.1 update

llama.cpp adds fa-vec tunings for Apple M3 Max, M5 and M5 Pro

llama.cpp b10665 adds DSpark support for Nemotron3.5