Alternatives and similar repositories for fastgen

Users that are interested in fastgen are comparing it to the libraries listed below

• tiiuae / onebitllms
  Lightweight toolkit package to train and fine-tune 1.58bit Language models
  ☆88Updated 3 months ago
• BlinkDL / modded-nanogpt-rwkv
  RWKV-7: Surpassing GPT
  ☆95Updated 9 months ago
• huggingface / kernel-builder
  👷 Build compute kernels
  ☆136Updated last week
• Zyphra / tree_attention
  Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters
  ☆129Updated 9 months ago
• ScalingIntelligence / good-kernels
  Samples of good AI generated CUDA kernels
  ☆89Updated 3 months ago
• UmerHA / triton_util
  Make triton easier
  ☆47Updated last year
• kroggen / mamba.c
  Inference of Mamba models in pure C
  ☆191Updated last year
• character-ai / pipelining-sft
  Simple and efficient DeepSeek V3 SFT using pipeline parallel and expert parallel, with both FP8 and BF16 trainings
  ☆81Updated last month
• Zyphra / Zamba2
  PyTorch implementation of models from the Zamba2 series.
  ☆185Updated 7 months ago
• schwartz-lab-NLP / TOVA
  Token Omission Via Attention
  ☆128Updated 11 months ago
• bloc97 / DeMo
  DeMo: Decoupled Momentum Optimization
  ☆190Updated 9 months ago
• opendatahub-io / vllm-tgis-adapter
  vLLM adapter for a TGIS-compatible gRPC server.
  ☆39Updated this week
• HazyResearch / lolcats
  Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"
  ☆245Updated 7 months ago
• JoeLi12345 / nGPT
  an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)
  ☆105Updated 6 months ago
• Cornell-RelaxML / yaqa-quantization
  ☆56Updated 2 months ago
• s-smits / grpo-optuna
  Optimizing Causal LMs through GRPO with weighted reward functions and automated hyperparameter tuning using Optuna
  ☆55Updated 7 months ago
• foundation-model-stack / bamba
  Train, tune, and infer Bamba model
  ☆131Updated 3 months ago
• NolanoOrg / SpectraSuite
  ☆51Updated last year
• PrimeIntellect-ai / pccl
  PCCL (Prime Collective Communications Library) implements fault tolerant collective communications over IP
  ☆120Updated this week
• HazyResearch / cartridges
  Storing long contexts in tiny caches with self-study
  ☆179Updated last week
• Mihaiii / backtrack_sampler
  An easy-to-understand framework for LLM samplers that rewind and revise generated tokens
  ☆146Updated 6 months ago
• Zyphra / Zyda_processing
  ☆39Updated last year
• apple / ml-recurrent-drafter
  ☆217Updated 7 months ago
• sekstini / basedxl
  ☆18Updated last year
• kyleliang919 / Super_Muon
  ☆63Updated 5 months ago
• chu-tianxiang / QuIP-for-all
  QuIP quantization
  ☆59Updated last year
• BorealisAI / neuzip
  Official repository for the paper "NeuZip: Memory-Efficient Training and Inference with Dynamic Compression of Neural Networks". This rep…
  ☆60Updated 10 months ago
• IST-DASLab / QuEST
  Work in progress.
  ☆72Updated 2 months ago
• Michaelvll / llm-ie-benchmarks
  A collection of reproducible inference engine benchmarks
  ☆33Updated 4 months ago
• astramind-ai / BitMat
  An efficent implementation of the method proposed in "The Era of 1-bit LLMs"
  ☆155Updated 11 months ago