mixture-of-experts

apache-2 ,  inference-engine ,  mixture-of-experts ,  kimi-k3 ,  quantization ,  local-llm ,  warp

cpu-inference ,  mixture-of-experts ,  kimi-k3 ,  kimi-k3-in-c ,  mxfp4 ,  c99 ,  memory-efficient

multimodal ,  open-weights ,  product-launch ,  tinker ,  inkling ,  thinking-machines ,  mixture-of-experts

mixture-of-experts ,  swift ,  local-llm ,  apple-silicon ,  metal ,  turbofieldfare ,  on-device-inference

agent ,  long-context ,  research ,  paper ,  reinforcement-learning ,  mixture-of-experts ,  llada ,  diffusion-llm

pretraining ,  opensource ,  long-context ,  mamba ,  paper ,  mixture-of-experts ,  llm

moonshot-ai ,  long-context ,  mixture-of-experts ,  moba ,  block-attention ,  sparse-attention ,  moonshotai ,  mixture-of-block-attention

product-launch ,  reinforcement-learning ,  mixture-of-experts ,  multimodal ,  open-weights ,  thinking-machines ,  inkling ,  tinker

quantization ,  cpu-inference ,  inference-engine ,  mixture-of-experts ,  speculative-decoding ,  local-llm ,  colibri

product-launch ,  mixture-of-experts ,  llm ,  apple ,  on-device ,  apple-intelligence ,  private-cloud-compute

mixture-of-experts ,  diffusion-language-model ,  multimodal ,  image-generation ,  llada ,  image-understanding ,  research ,  paper ,  diffusion-llm

pytorch ,  gpt ,  rlhf ,  mixture-of-experts ,  minimind ,  language-model ,  pretraining

apple-silicon ,  mixture-of-experts ,  llm-inference ,  quantization ,  on-device-ai ,  metal ,  flash-moe

mixture-of-experts ,  alibaba ,  sparse-moe ,  multimodal-agent ,  multi-token-prediction ,  alibaba-cloud ,  qwen3-5 ,  qwen3-5-397b-a17b ,  qwen3-5-plus ,  multimodal-native ,  gated-delta-networks

multimodal ,  vision-language ,  mixture-of-experts ,  alibaba ,  qwen ,  qwen3-vl

mixture-of-experts ,  dots-llm1 ,  rednote

mixture-of-experts ,  deepseek ,  deepep ,  expert-parallelism

mixture-of-experts ,  deepseek ,  expert-parallelism ,  expert-parallelism-load-balancer

microsoft ,  mixture-of-experts ,  small-llm ,  small-multimodal ,  phi-3 ,  phi-3-5

paper ,  mixture-of-experts ,  top-ml-papers-of-the-week ,  llm-reasoning ,  flashattention ,  routellm ,  flashattention-3 ,  rankrag ,  lookback-lens ,  internet-of-agents ,  3dgen ,  ttt-linear

nvidia ,  nvidia-h100 ,  mixture-of-experts ,  tensorrt-llm ,  tensorrt ,  mixtral-8x7b

together ,  paper ,  llm-framework ,  mixture-of-experts ,  mixture-of-depths ,  mixture-of-agents

mixture-of-experts ,  deepseek-v2 ,  deepseek ,  open-weights

geeknews ,  mistral-ai ,  mixture-of-experts ,  mistral ,  mixtral ,  mixtral-8x22b

paper ,  deepmind ,  mixture-of-experts ,  mod ,  mixture-of-depths ,  moe ,  mod-transformer ,  mode ,  expert-choice-mod-routing ,  mixture-of-depths-and-experts

mixture-of-experts ,  mamba ,  jamba ,  mamba-ssm ,  ai21labs ,  open-weights

mixture-of-experts ,  qwen ,  qwen-1-5-moe ,  qwen-moe ,  qwen-1-5-moe-a2-7b

geeknews ,  mixture-of-experts ,  long-context ,  gemini ,  google-gemini

multimodal ,  vision-language ,  mixture-of-experts ,  llava ,  moe-llava

paper ,  top-ml-papers-of-the-week ,  llm-survey ,  hallucination ,  mobile-aloha ,  llm-agent ,  multimodal ,  llm-finetuning ,  mixture-of-experts ,  gpt-4v ,  code-llm ,  llama-pro ,  llm-augment ,  docllm ,  instruct-imagen