--- id: "/infrastructure-chips-efficiency/advanced/" title: "Infrastructure, chips & efficiency for advanced learners" url: "https://skillissue.dev/infrastructure-chips-efficiency/advanced/" --- # Infrastructure, chips & efficiency for advanced learners [Beginner](/infrastructure-chips-efficiency/beginner/index.md) [Intermediate](/infrastructure-chips-efficiency/intermediate/index.md) [Advanced](/infrastructure-chips-efficiency/advanced/index.md) [Expert](/infrastructure-chips-efficiency/expert/index.md) 86 resources · page 1 of 3 - ### [Apple machine learning research](/model-training/training-infrastructure/apple-machine-learning-research/index.md) by [Apple Machine Learning Research](https://machinelearning.apple.com/) Apple Machine Learning Research publications: papers on speech, language, vision, privacy and health. entity org US; international en - ### [Apple MLX](/model-training/training-infrastructure/apple-mlx/index.md) by [ml-explore](https://github.com/ml-explore) Apple MLX GitHub repo: array framework for machine learning on Apple silicon. entity repo US en - ### [OpenInference](/model-training/training-infrastructure/arize-openinference/index.md) by [Arize AI](https://www.arize.com) Arize OpenInference GitHub repo: OpenTelemetry instrumentation and semantic conventions for LLM tracing. entity repo US en - ### [Truss](/model-training/training-infrastructure/baseten-truss/index.md) by [Baseten](https://baseten.co) Baseten Truss GitHub repo: packaging and serving ML models in production. entity repo US en - ### [checkpoint-engine](/model-training/training-infrastructure/checkpoint-engine/index.md) by [Moonshot AI](https://moonshot.ai) Moonshot checkpoint-engine GitHub repo: middleware for updating model weights in running LLM inference engines. entity repo China en - ### [Colossal-AI](/model-training/training-infrastructure/colossal-ai/index.md) by [HPC-AI Tech](https://hpc-ai.com/) HPC-AI Tech Colossal-AI GitHub repo: parallel training system for large AI models. entity repo China/Singapore en|zh - ### [DeepGEMM](/model-training/training-infrastructure/deepgemm/index.md) by [DeepSeek](https://www.deepseek.com/) DeepSeek DeepGEMM GitHub repo: GPU matrix multiplication kernel library including FP8 GEMMs. entity repo China en - ### [DeepGEMM-Ascend](/model-training/training-infrastructure/deepgemm-ascend/index.md) by [DeepSeek](https://www.deepseek.com/) DeepSeek DeepGEMM-Ascend GitHub repo: matrix multiplication kernels for Huawei Ascend hardware. entity repo China en - ### [DeepSeek Open Infra Index](/model-training/training-infrastructure/deepseek-open-infra-index/index.md) by [DeepSeek](https://www.deepseek.com/) DeepSeek Open Infra Index GitHub repo: index of released infrastructure tools such as FlashMLA, DeepEP, DeepGEMM and 3FS. entity leaderboard China en - ### [DeepSpeed](/model-training/training-infrastructure/deepspeed/index.md) by [deepspeedai](https://github.com/deepspeedai) DeepSpeed GitHub repo: deep learning optimisation library for large-scale training and inference, origin of ZeRO. entity repo US en - ### [py-why/dowhy](/model-training/training-infrastructure/dowhy/index.md) by [PyWhy](https://pywhy.org) PyWhy DoWhy GitHub repo: Python library for causal inference with explicit assumptions and refutation tests. entity repo International en - ### [Etched](/model-training/training-infrastructure/etched/index.md) by [Etched](https://etched.com/) Etched: company building transformer-specific AI inference chips, racks and software. entity lab US en - ### [FastDeploy](/model-training/training-infrastructure/fastdeploy/index.md) by [PaddlePaddle](https://www.paddlepaddle.org.cn/) PaddlePaddle FastDeploy GitHub repo: inference and deployment toolkit for LLMs and VLMs. entity repo China en|zh - ### [Fireworks cookbook](/model-training/training-infrastructure/fireworks-cookbook/index.md) by [Fireworks AI](https://fireworks.ai) Fireworks cookbook GitHub repo: recipes for fine-tuning, function calling and building with Fireworks AI. entity repo US en - ### [FlashMLA](/model-training/training-infrastructure/flashmla/index.md) by [DeepSeek](https://www.deepseek.com/) DeepSeek FlashMLA GitHub repo: multi-head latent attention decoding kernels for Hopper GPUs. entity repo China en - ### [FlashQLA](/model-training/training-infrastructure/flashqla/index.md) by [Qwen](https://qwen.ai/) QwenLM FlashQLA GitHub repo: linear attention kernel library built on TileLang. entity repo China en - ### [Huawei Ascend](/model-training/training-infrastructure/huawei-ascend/index.md) by [Huawei](https://www.huawei.com/) Huawei Ascend community site: Ascend AI processors, CANN software stack, frameworks and developer tools. entity lab CN zh - ### [KTransformers](/model-training/training-infrastructure/ktransformers/index.md) by [kvcache.ai](https://madsys.cs.tsinghua.edu.cn/) KTransformers GitHub repo: framework for heterogeneous CPU-GPU inference of large LLMs. entity repo CN en - ### [Kunlunxin (昆仑芯)](/model-training/training-infrastructure/kunlunxin/index.md) by [Kunlunxin (昆仑芯)](https://www.kunlunxin.com/) Kunlunxin site: Baidu-affiliated maker of AI accelerator chips (in Chinese). entity company China zh - ### [Liger Kernel](/model-training/training-infrastructure/liger-kernel/index.md) by [LinkedIn](https://engineering.linkedin.com) LinkedIn Liger Kernel GitHub repo: Triton kernels for LLM training. entity repo US en - ### [Liquid AI research](/model-training/training-infrastructure/liquid-ai-research/index.md) by [Liquid AI](https://liquid.ai/) Liquid AI research page: papers and technical reports on Liquid Foundation Models. entity org US en - ### [llm.c](/model-training/training-infrastructure/llm-c/index.md) by [Andrej](https://x.com/karpathy) llm.c GitHub repo (Andrej Karpathy): LLM training in plain C and CUDA. entity repo US en - ### [LMDeploy](/model-training/training-infrastructure/lmdeploy/index.md) by [InternLM](https://chat.intern-ai.org.cn/) InternLM LMDeploy GitHub repo: toolkit for compressing, deploying and serving LLMs. entity repo China en|zh - ### [llama.cpp](/model-training/training-infrastructure/local-on-device-ai/index.md) by [ggml](https://huggingface.co/ggml-org) llama.cpp GitHub repo: LLM inference in C/C++ for local and on-device hardware. entity repo Bulgaria en - ### [Megatron-LM](/model-training/training-infrastructure/megatron-lm/index.md) by [NVIDIA Corporation](https://nvidia.com) NVIDIA Megatron-LM GitHub repo: research code for training transformer models at scale with tensor and pipeline parallelism. entity repo US en - ### [MiniCPM](/model-training/training-infrastructure/minicpm-technical-repository/index.md) by [OpenBMB](https://www.openbmb.cn) OpenBMB MiniCPM GitHub repo: on-device small language models with deployment resources. entity repo China zh - ### [MiniCPM-V](/model-training/training-infrastructure/minicpm-v-technical-repository/index.md) by [OpenBMB](https://www.openbmb.cn) OpenBMB MiniCPM-V GitHub repo: on-device multimodal models MiniCPM-V and MiniCPM-o for vision and speech. entity repo China zh - ### [MLX / mlx-lm](/model-training/training-infrastructure/mlx-mlx-lm/index.md) by [ml-explore](https://github.com/ml-explore) Apple mlx-lm GitHub repo: Python package for running and fine-tuning LLMs with MLX. entity repo US en - ### [MobileCLIP research implementation](/model-training/training-infrastructure/mobileclip-research-implementation/index.md) by [Apple AI/ML Research](https://github.com/apple-aiml-research) Apple ml-mobileclip GitHub repo: official implementation of MobileCLIP efficient image-text models. entity repo US en - ### [Mooncake](/model-training/training-infrastructure/mooncake/index.md) by [kvcache.ai](https://madsys.cs.tsinghua.edu.cn/) Mooncake GitHub repo: KV-cache-centric disaggregated serving platform used for Kimi. entity repo China en - ### [Motif Technologies](/model-training/training-infrastructure/motif-technologies/index.md) by [Motif Technologies](https://github.com/MotifTechnologies) Motif Technologies GitHub organisation: training and serving code for its Korean models. entity org South Korea en - ### [Multiverse Computing model registry](/model-training/training-infrastructure/multiverse-computing-model-registry/index.md) by [Multiverse Computing](https://huggingface.co/MultiverseComputingCAI) Multiverse Computing Hugging Face organisation: compressed model cards such as HyperNova. entity org Spain / International en [Next page](/infrastructure-chips-efficiency/advanced/page/2/index.md) [Plain-text Markdown copy](/infrastructure-chips-efficiency/advanced/index.md)