Official implementation for Training LLMs with MXFP4
☆130Apr 25, 2025Updated last year
Alternatives and similar repositories for mxfp4-llm
Users that are interested in mxfp4-llm are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆129Mar 18, 2026Updated 7 months ago
- ☆53May 20, 2025Updated last year
- PyTorch emulation library for Microscaling (MX)-compatible data formats☆365Jul 17, 2026Updated 3 months ago
- ☆126Feb 26, 2026Updated 7 months ago
- Code for the papers: “Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling” and “Adaptive Block-Scaled Data Types”☆204Apr 21, 2026Updated 5 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Pytorch implementation of "Oscillation-Reduced MXFP4 Training for Vision Transformers" on DeiT Model Pre-training☆41May 4, 2026Updated 5 months ago
- QuTLASS: CUTLASS-Powered Quantized BLAS for Deep Learning☆204Sep 2, 2026Updated last month
- Ongoing research training transformer models at scale☆19Oct 9, 2026Updated last week
- Work in progress.☆82Nov 25, 2025Updated 10 months ago
- ☆96Jun 20, 2025Updated last year
- Best practices for training DeepSeek, Mixtral, Qwen and other MoE models using Megatron Core.☆204May 29, 2026Updated 4 months ago
- [ASPLOS'26] HILOS: A Cost-Effective Near-Storage Processing Solution for Offline Inference of Long-Context LLMs☆22Jan 18, 2026Updated 9 months ago
- LLM Inference with Microscaling Format☆35Nov 12, 2024Updated last year
- Boosting 4-bit inference kernels with 2:4 Sparsity☆101Sep 4, 2024Updated 2 years ago
- Simple, predictable pricing with DigitalOcean hosting • AdAlways know what you'll pay with monthly caps and flat pricing. Enterprise-grade infrastructure trusted by 600k+ customers.
- [ICML 2026]A framework to compare low-bit integer and float-point formats☆84May 6, 2026Updated 5 months ago
- HALO: Hadamard-Assisted Low-Precision Optimization and Training method for finetuning LLMs. 🚀 The official implementation of https://arx…☆32Feb 17, 2025Updated last year
- All-in-one repository for Fine-tuning & Pretraining (Large) Language Models☆15Mar 8, 2023Updated 3 years ago
- [ICML 2025] SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models☆67Aug 9, 2024Updated 2 years ago
- [PACT'24] GraNNDis. A fast and unified distributed graph neural network (GNN) training framework for both full-batch (full-graph) and min…☆10Aug 13, 2024Updated 2 years ago
- QuIP quantization☆66Mar 17, 2024Updated 2 years ago
- Quartet II Official Code☆83May 1, 2026Updated 5 months ago
- [ICML 2025] Official PyTorch implementation of "FlatQuant: Flatness Matters for LLM Quantization"☆229Nov 25, 2025Updated 10 months ago
- ☆28Jul 2, 2026Updated 3 months ago
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- ☆14May 3, 2022Updated 4 years ago
- Code for Neurips24 paper: QuaRot, an end-to-end 4-bit inference of large language models.☆535Nov 26, 2024Updated last year
- [DATE 2023] Pipe-BD: Pipelined Parallel Blockwise Distillation☆12Jul 13, 2023Updated 3 years ago
- Microsoft Automatic Mixed Precision Library☆634Dec 1, 2025Updated 10 months ago
- Fast Hadamard transform in CUDA, with a PyTorch interface☆354Mar 10, 2026Updated 7 months ago
- Intel Gaudi's Megatron DeepSpeed Large Language Models for training☆17Dec 19, 2024Updated last year
- Expert Specialization MoE Solution based on CUTLASS☆27Apr 14, 2026Updated 6 months ago
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,585Updated this week
- ☆35Mar 28, 2025Updated last year
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- PyTorch native quantization for training and inference☆3,005Updated this week
- ☆27May 9, 2025Updated last year
- Multiple GEMM operators are constructed with cutlass to support LLM inference.☆20Aug 3, 2025Updated last year
- [ACL 2026 Main] Code for the paper "ARCQuant: Boosting NVFP4 Quantization with Augmented Residual Channels for LLMs"☆33Sep 21, 2026Updated 3 weeks ago
- Official implementation of EMNLP'23 paper "Revisiting Block-based Quantisation: What is Important for Sub-8-bit LLM Inference?"☆24Oct 25, 2023Updated 2 years ago
- Faster Pytorch bitsandbytes 4bit fp4 nn.Linear ops☆30Mar 16, 2024Updated 2 years ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,164Sep 4, 2024Updated 2 years ago