China AI

AITinker announces that Z.ai's GLM-5.3 is now available on its platform with a 256k context window.

AIUnsloth quantized GLM-5.3 to a dynamic 1-bit version of 217GB, versus 1.5TB for BF16, retaining about 76% top-1 accuracy while cutting size by 83%.

AIUnsloth AI published a guide for running GLM-5.3 locally using quantized GGUF weights. The 2-bit version is reduced from 1.51TB to 239GB and retains about 81% accuracy, and it can run on a 256GB Mac or RAM/VRAM setups.

Why it matters: The guide shows which quantization levels fit local memory budgets and how much accuracy each costs, useful for planning a local deployment.

AILM Studio says Zhipu's GLM-5.3 is now available in its Bionic agent, hosted in the US with zero data retention. The launch offer is 50% off until Monday.

AILMSYS Org shared a link to a new SGLang cookbook page covering GLM-5.3 deployment. The post itself gives no further details on the model's features, performance, or setup steps.

AISGLang offers day-0 support for GLM-5.3 on NVIDIA Blackwell and Hopper and AMD MI300X, MI325X, and MI355X GPUs, using the same runtime and flags.

AIRadixArk has published an experimental NVFP4 checkpoint for Zhipu's GLM-5.3 on Hugging Face, and says Miles support for GLM-5.3 is on the way.

AIZ.ai has made GLM-5.3 open-weight, so users can download, run, and customize its weights. The company describes it as its most capable model for agentic coding and cyber defense, with weights on Hugging Face and details in a tech blog.

Why it matters: The source ties the open-weight release to coding, agentic, and cyber defense use, with a weights link for anyone who wants to run or customize it.

AIMeituan LongCat evaluated 7 frontier models on 36 AI R&D tasks covering 756 trajectories, looking beyond final scores.

Why it matters: The paper separates final scores from reliability and novelty, showing where agent research loops succeed and where they fall short.

AIRadixArk introduced LoRA SFT in Miles-diffusion for fast, targeted post-training of diffusion models.

AIThe Beijing Municipal Science and Technology Commission and Zhongguancun Management Committee named the Chinese Academy of Sciences' "Pan Shi" scientific foundation model among the first Beijing typical cases of AI empowering scientific research, announced at the 2026 Scientific Intelligence Conference.

AILMSYS Org posted a link to a blog post about MiniMax H3 on H200 hardware, dated August 27, 2026. The post itself offers no benchmark figures, specifications, or other details beyond the link.

AIMiniMax-H3 on 8×H200 GPUs reaches 1.85–1.95x lossless speedup over Diffusers without approximation, with fixed prompts, seeds, resolution, FPS, and 50 denoising steps.

AIGLM-5.3-Flash (ox-alpha) can be quantized to 4-bit while retaining 93% accuracy, according to Daniel Han.

AIUnsloth says GLM-5.3-Flash can run locally, with a 3-bit GGUF version needing 128GB of RAM and the 1-bit version working on 102GB of RAM or VRAM. The guide's table lists memory needs from 100GB at 1-bit to 650GB at BF16, and reports that the 1-bit quant keeps 71% of top-1% accuracy while being 85% smaller than BF16.

Why it matters: The guide gives concrete memory requirements for each quantization level, which helps readers judge whether the model fits their hardware.

AIOpenBMB released MiniCPM5-2B-SFT, an SFT-only BF16 checkpoint taken before RL and OPD, within its MiniCPM5-2B series.

Why it matters: The source gives concrete benchmark averages against same-size and larger models, plus released training data and multiple deployment formats, useful for judging a compact on-device model.

AIOpenBMB released MiniCPM5-2B, a dense 2B Transformer for on-device and resource-constrained deployment, alongside its training datasets.

AITencent's Hy Team released Hy4 preview, a Mixture-of-Experts model with 770B total parameters and 49B activated per token, with a 1M context length.

Why it matters: The model card gives architecture, activated parameters, and vLLM and SGLang deployment recipes, useful for judging whether the release fits your serving setup.

AIQwen has published Qwen-Drive-1.0-4B on Hugging Face, a vision-language model for autonomous driving built on Qwen3.5-4B. The release includes a BEV perception head and two Planning Experts, planner-sft and planner-rl, with code and an inference example in the linked GitHub repository.

Why it matters: The source gives concrete benchmark results and a runnable setup, letting readers judge how a driving VLM with planning and perception heads compares with existing systems.

AIZ.ai announced that the weights for GLM-5.3 will be released tomorrow. The post links to the GLM-5.3 model page on Hugging Face.

AITencent has published ContextPilot-E4B on Hugging Face, the Gemma4-E4B checkpoint of ContextPilot, a framework that teaches long-horizon language-model agents to plan, maintain long-term memory, and offload less useful context while reasoning and using tools.

AITencent has released ContextPilot-14B on Hugging Face, a Qwen3-14B checkpoint for proactive context management in long-horizon language-model agents.

AIReplicate has made Wan 3.0 Prime, the accelerated variant of Alibaba's Wan 3.0, available for text-to-video, image-to-video, and reference-driven workflows.

AILM Studio announced that Z.ai's GLM-5.3-Flash, previously previewed as Ox Alpha, is available in LM Studio Bionic.

AIZ.ai released GLM-5.3-Flash, a 320B-A18B model, with day-0 support in SGLang, after appearing earlier as ox-alpha.

Why it matters: The post reports GLM-5.3-Flash's native multimodal design, its efficiency claims, and day-0 SGLang support, which bear on running it in practice.

AIUnsloth announces that Qwen3.8-Flash-Next can be run locally through its GGUF quantizations. The source says the 1-bit version needs 75GB of RAM or unified memory, and that the 125B MoE model is reported to outperform Claude-Opus-4.6 (Max).

Why it matters: The source gives concrete local hardware requirements, quantization sizes, and a guide, showing how a 125B MoE model can run on a 75GB RAM setup.

AISGLang announced Day-0 support for Qwen3.8-Flash-Next, a 125B MoE model with 6B active parameters and 51B N-gram embeddings, in collaboration with Alibaba Qwen, NVIDIA, and AMD. The post reports 540 tok/s decode speed at BS=1 on NVIDIA B200 (TP4) with an NVFP4 checkpoint, and says N-gram host offloading saves 23.5 GiB VRAM per GPU and raises KV capacity by 78.5%.

AIZ.ai says GLM-5.3-Flash achieves greater intelligence with less compute, attributing this to architectural enhancements and an optimized pre-training corpus. The post gives no benchmark scores, parameter counts, or pricing.

AIOn the Z.ai Code Bench, which measures real-world coding performance, GLM-5.3-Flash clearly beats GLM-5.2 at every effort level. It performs on par with Claude Opus 4.8 on the same benchmark.

AIZ.ai has published standard API pricing for GLM-5.3-Flash, charging $0.15 per 1M input tokens and $0.50 per 1M output tokens. Cached input tokens are priced at $0.03 per 1M tokens.

AIZ.ai releases GLM-5.3-Flash, a 320B-A18B model with a 1M-token context window that accepts multimodal input.

Why it matters: The post gives the parameter count, license, context window, and access routes, which helps readers judge whether the model fits their own deployment needs.

AIAi2 has released Bwen-8B, a byte-level autoregressive language model retrofitted from Qwen3 8B Base through a short additional training procedure called byteification, which lets it operate over bytes instead of tokens.

AIAi2 has released Bwen-8B-Stage1 on Hugging Face, a byte-level autoregressive model retrofitted from Qwen3-8B-Base through a short additional training procedure.

AIDeepSeek V4 Pro 0813 scored 95.2% on SWE-Bench Verified, ahead of Kimi K3 at 92.6% and Fable 5 at 85.4%, in Fireworks AI's eval runs.

AIDeepSeek V4 Pro 0813 recorded zero refusals across 840 adversarial security tasks in CyberGym testing, solving them at about half the cost per success of the top-scoring model tested, Kimi K3.

AIGoogle Cloud has added native TPU support to vLLM and engineered optimizations to serve the Qwen3-Embedding-8B model on Cloud TPU, targeting 4K+ token text and 15K+ token multimodal inputs.

AIHarvey and Fireworks post-trained Tenet from the Kimi K3 base using asynchronous reinforcement learning on the Fireworks Training API for long-horizon legal work.

AIZ.ai has released GLM-5.3-Flash, a model with native visual capabilities that observe interfaces, rendering results, and interaction feedback across code, browsers, and GUIs.

Why it matters: The release notes give GLM-5.3-Flash's architecture, parameter counts, and cybersecurity findings, which make the model's scope concrete for comparison with earlier GLM releases.

AIDaniel Han says users can fine-tune Qwen3.8-27B for free on Kaggle with a Google account, which provides 30 hours of GPU time on 2× Tesla T4s.

AIUnsloth now lets users fine-tune Qwen3.8-27B for free with its notebook, with local training running on 24GB VRAM. The company says its method trains the model 1.5x faster while using 50% less VRAM.