AI projects from Hugging Face

DataWorld-35B-A3B

Zhanli-Li · Kit

A Qwen3.6-35B-A3B model tuned with multi-turn agentic reinforcement learning on the DataWorld tasks, for tool-calling agents. For teams serving a large mixture…

Qwen3.8-27B-QUASAR-NVFP4

QUASAR-QAT · Kit

Runs Qwen3.8-27B in 4-bit NVFP4 at 19.7 GB, trained with the QUASAR method to stay close to the full BF16 model. For people serving it with vLLM on an NVIDIA B…

GLM-5.3-Flash EXL3 4bpw for TensorFold

Mia-AiLab · Kit

4-bit EXL3 quantization of GLM-5.3-Flash prepared for TensorFold, with lower KL divergence than a popular TR3-4bpw quant at the same size. For people serving l…

DeepSeek-V4.1-Flash-1017-FLARE

local-inference-lab · Kit

Signed Flare runtime packages of DeepSeek V4.1 Flash 1017 for four-GPU serving, handling text and images behind an OpenAI-compatible API. For people running la…

Flint-DFlash2

incoai · Kit

A 403M-parameter draft model that speeds up serving of the Flint language model with speculative decoding in SGLang, without changing its output. It is not a s…

Qwen3.8-27B-NVFP4-DGX-SPARK

autotrust · Kit

Speeds up Qwen3.8-27B NVFP4 decoding about 3.9 times on a DGX Spark, using the same weights and a tuned speculative-decoding setup. For teams serving it with v…

gemma-4-31B-it-qat-GGUF

unsloth · Kit

Runs Google's Gemma 4 31B instruction model from Unsloth's 4-bit QAT GGUF files, with an included drafter for faster speculative decoding in llama.cpp. For peo…

Kimi-K3.1-Dflash2-Preview

RadixArk · Kit

Speeds up Kimi-K3.1-Pro serving by drafting blocks of tokens that the main model checks in one pass. A draft model for SGLang users, used alongside the target…

Ling-3.1-flash-DSpark

AQ-MedAI · Kit

A small draft model that speeds up Ling-3.1-Flash with speculative decoding in SGLang, drafting 5 or 7 tokens at a time. For teams serving Ling-3.1-Flash on NV…

Ling-3.1-flash-1019

inclusionAI · Kit

Handles scientific and agentic tasks with 560B total and 25B active parameters and a 1M-token context, tuned further from Ling-3.1-flash. Serving examples use…

Ling-3.1-flash

inclusionAI · Kit

Handles complex agent, search, office and coding work with 560B total and 25B active parameters and a 1M-token context. Serving examples use SGLang on multi-GP…

Qwen3.8-27B-DFlash2

z-lab · Kit

A draft model that speeds up Qwen3.8-27B by proposing a block of tokens for the main model to check. Not standalone. For people serving Qwen3.8-27B with SGLang…

Qwen3.8-27B-NVFP4

unsloth · Kit

A 4-bit NVFP4 copy of the Qwen3.8 27B vision-language model made by Unsloth, for serving with vLLM. For developers who want a dense model that reads images and…

Qwen3.8-27B-FP8

Qwen · Kit

Handles text, images and video with FP8-quantized weights of Qwen3.8-27B, a 27B dense model with adjustable thinking. For developers serving it with transforme…

sched-sim

prajwalpujari16 · Kit

A browser simulator for LLM serving schedulers with an animated Gantt chart that compares continuous batching, chunked prefill, shortest-job-first and disaggre…

RunsOnMine

prajwalpujari16 · Kit

A browser registry of LLM serving benchmarks measured on consumer GPUs, dual Tesla T4 and an RTX 3050 laptop, with concurrency scaling curves and a way to subm…

mini-infer

prajwalpujari16 · Kit

A browser studio that simulates GPU PagedAttention block allocation, prefix caching, speculative decoding checks and KV-cache sizing. For engineers and student…

infersize

prajwalpujari16 · Kit

A browser tool that estimates the GPU memory an LLM needs, compares GPUs, sizes speculative decoding latency and estimates cloud cost per million tokens, befor…