LiquidAI/LFM2.5-24B-A2B-DSpark开源了新模型
根据官方来源,LiquidAI/LFM2.5-24B-A2B-DSpark开源了新模型。详细信息请以原始来源为准。
证据来源
查看来源摘录
Liquid AI LiquidAI/LFM2.5-24B-A2B-DSpark LiquidAI/LFM2.5-24B-A2B-DSpark https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark
查看来源摘录
LiquidAI/LFM2.5-24B-A2B-DSpark · Hugging Face Hugging Face (https://huggingface.co/) Models (https://huggingface.co/models) Datasets (https://huggingface.co/datasets) Spaces (https://huggingface.co/spaces) Buckets new (https://huggingface.co/storage) Docs (https://huggingface.co/docs) Enterprise (https://huggingface.co/enterprise) Pricing (https://huggingface.co/pricing) Website Tasks (https://huggingface.co/tasks) HuggingChat (https://huggingface.co/chat) Collections (https://huggingface.co/collections) Languages (https://huggingface.co/languages) Organizations (https://huggingface.co/organizations) Community Blog (https://huggingface.co/blog) Posts (https://huggingface.co/posts) Daily Papers (https://huggingface.co/papers) Hardware (https://huggingface.co/hardware) Learn (https://huggingface.co/learn) Discord (https://huggingface.co/join/discord) Forum (https://discuss.huggingface.co/) GitHub (https://github.com/huggingface) Solutions Team & Enterprise (https://huggingface.co/enterprise) Hugging Face PRO (https://huggingface.co/pro) Enterprise Support (https://huggingface.co/support) Inference Providers (https://huggingface.co/inference/models) Inference Endpoints (https://huggingface.co/inference-endpoints) Storage Buckets (https://huggingface.co/storage) Log In (https://huggingface.co/login) Sign Up (https://huggingface.co/join) (https://huggingface.co/LiquidAI) LiquidAI (https://huggingface.co/LiquidAI) / LFM2.5-24B-A2B-DSpark (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark) Like 5 Follow Liquid AI 6.19k Text Generation (https://huggingface.co/models?pipeline_tag=text-generation) Safetensors (https://huggingface.co/models?library=safetensors) sglang (https://huggingface.co/models?other=sglang) qwen3 (https://huggingface.co/models?other=qwen3) speculative-decoding (https://huggingface.co/models?other=speculative-decoding) dspark (https://huggingface.co/models?other=dspark) lfm2 (https://huggingface.co/models?other=lfm2) lfm2_moe (https://huggingface.co/models?other=lfm2_moe) draft-model (https://huggingface.co/models?other=draft-model) Mixture of Experts (https://huggingface.co/models?other=moe) License: lfm1.0 Model card (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark) Files Files and versions xet (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark/tree/main) Community (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark/discussions) Copy to bucket new LFM2.5-24B-A2B-DSpark (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark#lfm25-24b-a2b-dspark) 🗒️ Model Details (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark#🗒️-model-details) 📊 Performance (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark#📊-performance) 🏃 How to run (SGLang) (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark#🏃-how-to-run-sglang) 📬 Contact (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark#📬-contact) Citation (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark#citation) Try LFM (https://playground.liquid.ai/) • Docs (https://docs.liquid.ai/lfm/getting-started/welcome) • Discord (https://discord.com/invite/liquid-ai) (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark#lfm25-24b-a2b-dspark) LFM2.5-24B-A2B-DSpark LFM2.5-24B-A2B-DSpark is a speculative-decoding draft model for LiquidAI/LFM2.5-24B-A2B (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B) . It accelerates generation by proposing tokens that the target model verifies. On NVIDIA B300 with the FP8 target in SGLang, output throughput averages 2.4× the target-only baseline at concurrency 1 (up to 3.1×), and 1.7× at concurrency 256. Find more information about LFM2.5-DSpark in our blog post (https://www.liquid.ai/blog/lfm2.5-dspark) . (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark#🗒️-model-details) 🗒️ Model Details Target model: LiquidAI/LFM2.5-24B-A2B (https://huggingface.co/LiquidAI/LFM2.5-24B-A2B) Draft parameters: 327.7M (BF16) Backbone: 5 full attention layers, hidden_size=2048, i
来自 星盘大模型百科