56 posts · How-to, Explainer, Model deep dive, Tutorial, Tip, Tool, Roundup
A 3B-decoder streaming ASR model with 240–560 ms delay. With the authors' adapted vLLM build, a rolling KV cache lets it transcribe audio of any length at constant memory.
Edge0/Audio8-ASR-Infinite
Tencent's MIT-licensed AuK does zero-shot TTS, speech editing, enhancement, and separation from natural-language instructions.
tencent/AuK
Breeze TTS 2 is an open-weight English and Chinese TTS model with voice cloning, text-described voices and a streaming API. Weights are non-commercial.
BreezeBlue/Breeze-TTS-2
Why AI detectors often get it wrong, which clues are worth noticing, and how content labels and watermarks fit in.
A contrastive scorer on frozen Qwen3-8B embeddings. It ranks the candidates you give it and answers typed questions about a state.
Contrastive-LM/CLM-v0.1-8B
NetEase Youdao's streaming ASR model built on Qwen3-ASR: transcript text never gets rewritten, with 80 ms to 2 s chunks. Setup, code and caveats.
netease-youdao/Confucius4-R2T2
DeepSeek's 552B MoE activates 8B params on prefill and stores 890 bytes of KV cache per token. What the card says and how to start running it.
deepseek-ai/DeepSeek-V4.1-Flash
Edge0-35B-A3B streams experts from SSD so a 4-bit Qwen3.6-35B-A3B derivative decodes at about 15 tok/s in under 3 GiB of active memory.
Edge0/Edge0-35B-A3B-preview
Google DeepMind's largest dense Gemma 4 model handles images, video frames and 256K-token context. Here's how to run it with Transformers.
google/gemma-4-31B-it
Get AI to point out real problems in your writing, judge it against your own checklist, and suggest fixes without replacing your voice.
A LoRA adapter plus a small decision head on Gemma-4-26B-A4B that gives a calibrated probability for every option in about 45 ms, and can think when it is unsure.
autotrust/GEV-26B-Decide
A 340M classifier from Fastino that scores intent, urgency, sentiment and routing labels you choose at call time, in one forward pass, on CPU or GPU.
fastino/GLiNER2.5-Decide
Z.ai's first natively multimodal GLM-5 model: 320B total / 18B active parameters, hybrid sparse-linear attention, MIT license. What the card covers.
zai-org/GLM-5.3-Flash
A plain-English look at how AI turns a sentence into a picture, why it sometimes gets things wrong, and where the copyright debate stands.
Skip the spec sheets. Learn four simple questions to ask, from privacy to free vs paid, and an easy side-by-side test to find the right AI assistant.
Use four simple parts (task, context, example, format) to turn vague AI answers into ones you can actually use. Includes before-and-after prompts.
A 27B vision decision model that returns a calibrated probability for every option in one forward pass. Here's how to serve it and use it.
autotrust/JEV-27B-VL
Jina-OCR-v1 is a 570M-active-parameter OCR model that turns page images into Markdown. Here is how to run it with Transformers, vLLM, or the hosted API.
jinaai/jina-ocr-v1
IFM's open MoE model runs 4B active parameters with a 512K context window. How to serve it, call it, and use it for agents and long documents.
IFM/K2-Horizon-MoVA-36B-A4B
Kimi K3 is a 2.8T-parameter open-weight multimodal MoE with a 1M-token context. What the model card says and how to call it.
moonshotai/Kimi-K3
Kokoro-82M is an 82M-parameter open-weight TTS model. How to install it, generate speech, and use it for narration, pronunciation fixes, and captions.
hexgrad/Kokoro-82M
Laya answers typed questions about text with probabilities instead of generated text. It covers 100+ languages and is Apache 2.0. Fit temperatures before trusting the probabilities.
convaiinnovations/laya
Xiaomi's 9B SFT model, fine-tuned from Qwen3.5-9B, scores higher than its base on SWE Pro, Terminal Bench and Toolathlon. How to serve it with SGLang.
XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B
Xiaomi's 1.02T-parameter MoE model has 1M context and omnimodal input, and was trained for agents. Here is how to serve it with vLLM or SGLang.
XiaomiMiMo/MiMo-V2.6-Pro-RL
MiniCPM5-2B is a 2.5B Llama-architecture model with 131K context and tool calling. Here's how to serve it and what it's good for.
openbmb/MiniCPM5-2B
MiniMax H3 is an open-weight model that generates video and stereo audio together. Here's how to serve it with SGLang and where the hosted APIs fit in.
MiniMaxAI/MiniMax-H3
NVIDIA's 100M-parameter diarizer labels up to 8 speakers, streaming with 0.32 s of buffer or offline with 30.4 s. Here's how to run it.
nvidia/Nemotron-3-Diarization
NeoHorse-1-4B is a text-only Qwen3.5-4B fine-tune for agent harnesses, tool use and coding. Here is how to serve it with SGLang or vLLM.
TokenRhythm/NeoHorse-1-4B
Nex-AGI's smallest Nex-N2.5 agent model, served with SGLang on 2 H100s, with tool calling and per-request thinking modes.
nex-agi/Nex-N2.5-mini
Learn what makes an AI model 'open' or 'closed', and how that choice affects your privacy, your costs, and where the AI can run.
Use any AI assistant to sketch a realistic travel plan, then check the hours, travel times and prices yourself before you book anything.
Use any AI assistant to plan a week of affordable meals around your schedule and tastes, with a shopping list and smart use of leftovers.
Qwen's driving model adds a BEV perception head and a flow-matching planner to Qwen3.5-4B. Install it and sample ego trajectories from demo scenes.
Qwen/Qwen-Drive-1.0-4B
Qwen-Image-2.1 combines text-to-image, image editing and native RGBA output in one diffusers pipeline. Here is how to run it.
Qwen/Qwen-Image-2.1
Qwen3.8-27B is an Apache-2.0 dense 27B vision-language model with adjustable reasoning. How to serve it and call it today.
Qwen/Qwen3.8-27B
Qwen's 125B MoE with 6B active params handles text, images and video. What it's good at, how to call it, and where it falls short.
Qwen/Qwen3.8-Flash-Next
Use an AI assistant to quiz yourself, explain hard topics simply, and check your answers, while still doing the learning yourself.
Get a clear summary of a long report, contract or PDF from an AI assistant, then check it against the original so you can trust it.
TeleOCR is a 1.2B Apache-2.0 model that turns text, tables, formulas and layouts into structured output, including from photos of warped pages.
XingChen-AGI/TeleOCR
A plain-English guide to tokens and context windows, and why long AI chats lose track of earlier details or reject a long file.
Use any AI assistant to pull owners, deadlines and next steps out of scrappy meeting notes, then draft the follow-up email.
Use an AI assistant to turn a confusing letter from your bank, doctor or landlord into plain English, safely, and know what to ask next.
Turn an AI assistant into a tutor that explains things at your pace, quizzes you, and checks that you really understand.
Use an AI assistant to find likely questions, shape your stories, and run a mock interview, without memorizing a script.
Teach any AI assistant your writing style with a few samples and simple tone instructions, then edit so every email still sounds like you.
A plain-English guide to AI agents: how they differ from chatbots, what tools they use, and where they're useful or still unreliable.
A plain-English guide to the technology behind AI chatbots: how they learn, why they sound so sure of themselves, and where they fall short.
A plain guide to keeping passwords, personal details, client data and health information out of AI chats, plus safer ways to get the same help.
AI assistants sometimes state false things with total confidence. Here's why it happens and a few simple habits that help you spot it.
China Telecom's 29B MoE activates 4B parameters per token and has a 256K context. Here's how to call it through an OpenAI-compatible API.
XingChen-AGI/Xing4.0-29B-A4B
YuE2-3B turns lyrics and a style prompt into a full 48 kHz stereo song. You can edit the melody and chords as an ABC score and render it again.
m-a-p/YuE2-3B
Clef answers typed questions about any input (text, JSON, images) with a probability per option in one forward pass. Here's how to use it for routing, triage and classification.
Cloudflare/clef
Google's new 740M multimodal embedding model runs on a CPU. Here's how to build a working semantic search over your own documents, and the prefixes that make it accurate.
google/embeddinggemma-2
Loading EmbeddingGemma 2 in float16 silently breaks it. Here's how to detect it and pick the right dtype for your hardware.
google/embeddinggemma-2
Kolibri 1 is a German–English reasoning model with only 3.5B active parameters. How to serve it, control its thinking effort, and wire up tool calling.
Aleph-Alpha/Kolibri-1
A practical rule of thumb for truncating embeddings to cut vector-DB cost, with the numbers from EmbeddingGemma 2.