philipturner / metal-flash-attentionLinks
FlashAttention (Metal Port)
☆535Updated last year
Alternatives and similar repositories for metal-flash-attention
Users that are interested in metal-flash-attention are comparing it to the libraries listed below
Sorting:
- Efficient framework-agnostic data loading☆438Updated this week
- Benchmark of Apple MLX operations on all Apple Silicon chips (GPU, CPU) + MPS and CUDA.☆199Updated 3 months ago
- ☆217Updated 8 months ago
- ☆560Updated 11 months ago
- Run transformers (incl. LLMs) on the Apple Neural Engine.☆63Updated last year
- CLI to demonstrate running a large language model (LLM) on Apple Neural Engine.☆116Updated 9 months ago
- 1.58 Bit LLM on Apple Silicon using MLX☆223Updated last year
- Official implementation of Half-Quadratic Quantization (HQQ)☆877Updated 3 weeks ago
- Python bindings for ggml☆146Updated last year
- Large Language Models (LLMs) applications and tools running on Apple Silicon in real-time with Apple MLX.☆455Updated 8 months ago
- Fast parallel LLM inference for MLX☆220Updated last year
- C API for MLX☆134Updated this week
- On-device Image Generation for Apple Silicon☆657Updated 5 months ago
- LLM training in simple, raw C/Metal Shading Language☆58Updated last year
- Code for the paper "QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models".☆277Updated last year
- Inference of Mamba models in pure C☆192Updated last year
- Convert StableHLO models into Apple Core ML format☆19Updated 2 months ago
- LLM-based code completion engine☆191Updated 8 months ago
- Apple GPU microarchitecture☆550Updated last year
- Code for paper: "QuIP: 2-Bit Quantization of Large Language Models With Guarantees"☆383Updated last year
- Fast low-bit matmul kernels in Triton☆373Updated last week
- An innovative library for efficient LLM inference via low-bit quantization☆348Updated last year
- LM Studio Apple MLX engine☆790Updated this week
- ModernBERT model optimized for Apple Neural Engine.☆27Updated 8 months ago
- An efficent implementation of the method proposed in "The Era of 1-bit LLMs"☆155Updated 11 months ago
- FastMLX is a high performance production ready API to host MLX models.☆331Updated 6 months ago
- ☆55Updated 2 years ago
- CLIP inference in plain C/C++ with no extra dependencies☆522Updated 3 months ago
- scalable and robust tree-based speculative decoding algorithm☆359Updated 8 months ago
- A high-throughput and memory-efficient inference and serving engine for LLMs☆266Updated 11 months ago