New in Inference
The inference projects, models and papers gaining momentum, newest and fastest first. Updated every two hours.
| Signal | Peak day | Total |
|---|---|---|
| Repo · deepseek-ai/DeepInferNew · 7d oldDeepInfer delivers high-throughput inference for DeepSeek modelsDeepInfer is an open-source Python framework that runs DeepSeek-V4.1-Flash models at production speed. It bundles optimized kernels, parallel attention and a zero-overhead batch scheduler so you can serve large language models offline without hand-crafting low-level code. | ▲ 2.9k | 20.4k |
| Repo · wheresryan22/anatomyNew · 15d oldwheresryan22/anatomy | ▲ 1.3k | 19.4k |
| Model · abenzerps/Qwen-Image-2.1-Uncensored-Quantizedabenzerps/Qwen-Image-2.1-Uncensored-Quantized | ▲ 147 | 4.9k |
| Model · Qwen/Qwen3.8-27BQwen releases 27B vision-language model, boosting image-to-text AIQwen3.8-27B is a 27-billion-parameter model that can look at images or video frames and generate descriptive text. It also supports complex multi-step reasoning and autonomous planning, making it useful for coding help, research, and long-horizon tasks. | 18.2k | |
| Model · Qwen/Qwen3.8-Flash-NextQwen3.8-Flash-Next offers faster long-context LLM inferenceQwen3.8-Flash-Next is an open-weight language model that can handle up to a million tokens while keeping latency low, thanks to its new hybrid attention called Qwen Sparse Attention. It runs with the standard Hugging Face Transformers stack and is compatible with vLLM, SGLang and TokenSpeed. | 6.4k | |
| Model · unsloth/Qwen3.8-27B-GGUFUnsloth releases 4-bit Qwen3.8-27B GGUF for desktop AIThe Qwen3.8-27B GGUF is a 27-billion-parameter model quantized to 4-bit, runnable on a laptop through Unsloth Desktop. It handles text, images and video, supports tool calling and can be fine-tuned for coding or long-horizon agentic tasks. | 5.3k | |
| Model · baidu/Unlimited-OCRbaidu/Unlimited-OCR | 4.9k | |
| Lab · Announcing new ways to customize models with SageMaker AI, without managing GPUsAnnouncing new ways to customize models with SageMaker AI, without managing GPUs | ||
| Lab · Write your own training loop, without managing GPUs, with Amazon SageMaker training session APIsWrite your own training loop, without managing GPUs, with Amazon SageMaker training session APIs | ||
| Paper · Making MLLM Latents Matter: Factor-Aware Predictive Learning for Cinematic Video GenerationNew Method Lets MLLMs Generate Cinematic Videos from Sparse PromptsIt uses factor-aware predictive learning to turn brief textual or sketch inputs into full-length, movie-style video clips. The system separately controls characters, scenes, actions and camera while automatically filling in the missing frames. | 100 | |
| Paper · Deep Fusion Memory: A Chunk-Level Parametric Memory for Large Language ModelsChunk-Level Parametric Memory Lets LLMs Learn Without ForgettingDeep Fusion Memory adds a trainable chunk-level store that LLMs can write to and read from during inference, sidestepping costly external retrieval. The memory updates with new data while preserving what the model already knows. | 32 |