abacusai / gh200-llm
Packages and instructions for training and inference of LLMs on NVIDIA's new GH200 machines
☆16Updated last month
Related projects ⓘ
Alternatives and complementary repositories for gh200-llm
- ☆96Updated last month
- Code for the paper "QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models".☆261Updated last year
- experiments with inference on llama☆105Updated 5 months ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆104Updated last month
- ring-attention experiments☆96Updated 3 weeks ago
- 🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash…☆191Updated 3 weeks ago
- Manage scalable open LLM inference endpoints in Slurm clusters☆236Updated 4 months ago
- An efficent implementation of the method proposed in "The Era of 1-bit LLMs"☆154Updated 3 weeks ago
- vLLM: A high-throughput and memory-efficient inference and serving engine for LLMs☆89Updated this week
- Cold Compress is a hackable, lightweight, and open-source toolkit for creating and benchmarking cache compression methods built on top of…☆86Updated 3 months ago
- Applied AI experiments and examples for PyTorch☆160Updated last week
- A toolkit for fine-tuning, inferencing, and evaluating GreenBitAI's LLMs.☆73Updated 3 weeks ago
- ☆91Updated last year
- Simple implementation of Speculative Sampling in NumPy for GPT-2.☆89Updated last year
- A pipeline for LLM knowledge distillation☆77Updated 3 months ago
- Simple and efficient pytorch-native transformer training and inference (batched)☆61Updated 7 months ago
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆216Updated 7 months ago
- Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆70Updated last week
- ☆72Updated 4 months ago
- Experiment of using Tangent to autodiff triton☆72Updated 9 months ago
- Benchmark suite for LLMs from Fireworks.ai☆58Updated this week
- Cataloging released Triton kernels.☆133Updated 2 months ago
- Archon provides a modular framework for combining different inference-time techniques and LMs with just a JSON config file.☆119Updated 3 weeks ago
- Collection of autoregressive model implementation☆66Updated last week
- Long context evaluation for large language models☆187Updated last week
- Simple and fast low-bit matmul kernels in CUDA / Triton☆140Updated this week
- ☆35Updated 2 weeks ago
- A safetensors extension to efficiently store sparse quantized tensors on disk☆46Updated this week
- Custom kernels in Triton language for accelerating LLMs☆17Updated 7 months ago
- Comprehensive analysis of difference in performance of QLora, Lora, and Full Finetunes.☆81Updated last year