Skip to content
Only Models#China

#China

Oct 8

TodayOct 8Thu1 item
  1. PandailyAI score57

    Shanghai AI Lab Open-Sources Intern-Decision Small Models for Structured Decisions

    Shanghai AI Lab has open-sourced Intern-Decision, a family of 0.8B, 2B and 4B parameter models that return structured decisions with probabilities instead of free text. The developers self-report that the 4B model averages 90.02% accuracy across seven test suites, ahead of a commercial reference model at 88.74%, with about 44 milliseconds of local latency on a single RTX 4090. Weights are on Hugging Face, and MetaX says the models run on its hardware from launch.

Oct 7

Oct 7Wed
  1. IThome · AI (IT之家)AI score72

    Anthropic releases Claude Haiku 5.5, cutting run costs about 75% from Haiku 4.5

    Anthropic released Claude Haiku 5.5, which it calls the fastest, cheapest, and most capable Haiku model so far. On average it costs about 75% less to run than Haiku 4.5, with input at $0.10 and output at $0.50 per million tokens for requests up to 100,000 tokens. Anthropic also cut Sonnet 5.5's cache read price from $0.20 to $0.10 per million tokens, which it says lowers run costs by about 20% on many agent tasks.

  2. Hugging Face BlogAI score53

    TII releases Falcon-ASR, a 1.6B speech recognition model focused on Emirati Arabic

    The Technology Innovation Institute introduces Falcon-ASR, a 1.6 billion parameter speech recognition model for Arabic with a focus on the Emirati dialect. On six Arabic test sets it reports an average word error rate of 20.92%, versus 23.17% for the best published leaderboard result it compared against. The model also transcribes English, French, Spanish and Portuguese with the same weights, and a demo Space is available while API access and native apps are planned.

Oct 5

Oct 5Mon
  1. IThome · AI (IT之家)AI score49

    Reflection AI releases open-weight Beam model to rival DeepSeek and Kimi

    Reflection AI, an Nvidia-backed startup, released Beam, its first open-weight large model, aimed at coding and agent tasks. The company says Beam is comparable to Z.ai's GLM-5.2 and is approaching Qwen3.8-Max on coding and agent work. Beam has 501 billion total parameters, with 23 billion activated per task in a sparse architecture.

Sep 23

Sep 23Wed
  2. ModelScopeAI score62

    Xiaomi MiMo-V2.6 open-sourced as a multimodal agent model family under MIT License

    Xiaomi has released MiMo-V2.6 as an open model family under the MIT License, designed for large-scale reinforcement learning. MiMo-V2.6-Pro scores 46 on the Artificial Analysis Intelligence Index, with 71.9 on DeepSWE v1.1, 89.9 on Terminal-Bench 2.1, and 82.0 on OSWorld-Verified. The 1.02T-parameter MoE activates 42B parameters and supports text, image, video, and audio input with a 1M-token context.

Sep 21

Sep 21Mon
  1. Xiaomi MiMoAI score38

    More intelligence. Same price. 📈 MiMo-V2.6 pushes the intelligence–cost Pareto frontier outward once again. 🔹 API pricing unchanged from V2.5, for both Pro and Flash 🔹 MiMo-V2.6-Pro sets a new price-performance record among Chinese models 🔹 At comparable intelligence, Pro costs just 1/20 to 1/60 as much as leading international models

    More intelligence. Same price. 📈 MiMo-V2.6 pushes the intelligence–cost Pareto frontier outward once again. 🔹 API pricing unchanged from V2.5, for both Pro and Flash 🔹 MiMo-V2.6-Pro sets a new price-performance record among Chinese models 🔹 At comparable intelligence, Pro costs just 1/20 to 1/60 as much as leading international models

Sep 19

Sep 19Sat
  1. StepFunAI score62

    StepFun Launches Step 5 Preview, a 600B MoE Model for Agentic Work

    StepFun has released Step 5 Preview, a flagship model for agentic work that it says delivers frontier-level performance in software engineering and professional knowledge work, with particular strength in finance. The model is a 600B total, 27B active mixture-of-experts design with a 1M context window and vision support. StepFun says it offers substantially lower task cost at comparable intelligence, and open weights are scheduled for October 15.

Sep 12

Sep 12Sat
  1. InternLM (Shanghai AI Lab) · new models on Hugging FaceAI score58

    Shanghai AI Lab releases Intern-S2-397B, a 397B multimodal scientific model

    Shanghai AI Lab's InternLM team released Intern-S2-397B, a multimodal foundation model for scientific intelligence and long-horizon agents. The model uses visual pre-training on raw scientific literature pages, multi-task reinforcement learning across more than 20 scientific domains, and agentic reinforcement learning in sandboxed environments.

Sep 11

Sep 11Fri
  1. BAAIAI score46

    Introducing AREX. A research agent from @BAAIBeijing . It does not treat a hard question as one long search. It drafts a candidate, checks every constraint, and goes back for what is still open. 122B MoE. 10B active. On hard search benches it sits next to GPT-5.4. Here is how it works👇

    Introducing AREX. A research agent from @BAAIBeijing . It does not treat a hard question as one long search. It drafts a candidate, checks every constraint, and goes back for what is still open. 122B MoE. 10B active. On hard search benches it sits next to GPT-5.4. Here is how it works👇

Sep 10

Sep 10Thu
  1. DeepSeekAI score46

    💾 Smaller KV cache. Bigger savings. Compared with the previous generation, V4.1-Flash’s KV cache needs just: 🔹 1/4 the HBM 🔹 1/8 the SSD storage Cache-hit charges often account for a large share of agent costs. Compressing the cache cuts those costs significantly. 3/6

    💾 Smaller KV cache. Bigger savings. Compared with the previous generation, V4.1-Flash’s KV cache needs just: 🔹 1/4 the HBM 🔹 1/8 the SSD storage Cache-hit charges often account for a large share of agent costs. Compressing the cache cuts those costs significantly. 3/6

Sep 4

Sep 4Fri
  1. BAAI · new models on Hugging FaceAI score26

    ConsiSpace: BAAI and Peking University release geometry-consistent video spatial reasoning model

    BAAI and Peking University researchers released official weights for ConsiSpace, a geometry-consistent multimodal framework for spatial reasoning in long-form visual observations. The model is described in the paper "ConsiSpace: Learning Geometric Consistency Matters for Video Spatial Reasoning" (arXiv:2607.17599).

Sep 3

Sep 3Thu

Aug 26

Aug 26Wed
  1. Unsloth AIAI score78

    Unsloth explains how to run Qwen3.8-Flash-Next locally on 75GB RAM

    Unsloth 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).

    AIWhy 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.

Aug 25

Aug 25Tue
  1. Z.ai (GLM) · new models on Hugging FaceAI score72

    Z.ai releases GLM-5.3-Flash, a natively multimodal model with 320B parameters

    Z.ai released GLM-5.3-Flash on Hugging Face, the first natively multimodal model in the GLM-5 series, with 320B total parameters and 18B active parameters. The source says it outperforms GLM-5.2 across benchmarks at one-tenth the price and approaches Claude Opus 4.8 on coding and agentic benchmarks. It adopts a hybrid sparse and linear attention architecture to reduce long-context serving costs.

    AIWhy it matters: The release shows a hybrid sparse and linear attention design aimed at cutting long-context serving costs, which is useful for comparing efficiency trade-offs.

  2. Z.ai (GLM) · new models on Hugging FaceAI score72

    Z.ai releases GLM-5.3 open weights with gains from post-training

    Z.ai released GLM-5.3 on Hugging Face, built on the same base model as GLM-5.2, with all gains coming from post-training. The source reports a 50% improvement over GLM-5.2 on Z.ai Code Bench and open-source SOTA on Terminal Bench 3.0 and Agents' Last Exam, with a benchmark table comparing it against Kimi K3, DeepSeek-V4 Pro-0813, Qwen3.8-Max, and others.

    AIWhy it matters: The source gives benchmark tables against GLM-5.2 and rival models, showing where the post-training gains concentrate in coding and cyber tasks.

Jul 27

Jul 27Mon
  1. KimiAI score86

    Moonshot AI releases Kimi K3 weights and technical report

    Moonshot AI is releasing the model weights and technical report for Kimi K3, a 2.8T-parameter MoE model with native visual understanding and a 1M-token context window. The post says the new architecture delivers 2.5x the intelligence per unit of compute, and the company is also opening high-performance attention kernels, an MoE communication library, and infrastructure for running agent environments at scale.

    AIWhy it matters: The source names the model size, context window, and released weights, which helps readers compare its scale and openness with other frontier releases.

Jul 23

Jul 23Thu
  1. BAAI · new models on Hugging FaceAI score62

    BAAI releases AREX-Base, a 122B deep research agent model

    BAAI has released AREX-Base, a 122B-total, 10B-activated Mixture-of-Experts deep research agent built on Qwen3.5-122B-A10B with a 262,144-token context. The model uses an inner research loop and an outer self-improvement loop, and the source reports it scoring 82.5 on BrowseComp and 85.4 on GAIA, under Apache 2.0.

    AIWhy it matters: The release pairs a 122B-parameter deep research agent with benchmark tables against frontier and open models, letting readers compare its search-agent results directly.

  2. BAAI · new models on Hugging FaceAI score47

    BAAI releases AREX-Turbo, a compact 4B recursive self-improving deep research agent

    BAAI's AREX-Turbo is a dense 4B deep research agent built on Qwen3.5-4B with a 262,144-token context length. It scores 70.7 on BrowseComp, 81.6 on GAIA and 40.6 on HLE with tools, versus 82.5, 85.4 and 52.4 for the 122B AREX-Base. The model is released under Apache License 2.0 and targets lower-cost research-agent deployment.

Jul 16

Jul 16Thu
  1. Soumith ChintalaAI score60

    Kimi K3 launches as a 2.8 trillion parameter open-weight model

    Moonshot AI announced Kimi K3, a native multimodal model with 2.8 trillion parameters and a 1 million token context window. The announcement cites up to 6.3x faster decoding in million-token contexts and about 25% higher training efficiency, and says open weights arrive by July 27, 2026. The author, Soumith Chintala, reposted it with a brief note of congratulations.

Jun 15

Jun 15Mon
  1. Z.ai Release NotesAI score62

    Z.ai Release Notes: GLM-5.2 Adds 1M Lossless Context for Long Tasks

    Z.ai's release notes list GLM-5.2 as supporting 1M lossless context, with improved long-horizon task performance and reduced context drift and goal forgetting. The company says GLM-5.2 achieves open-source SOTA performance on coding and long-horizon task benchmarks. The page also includes the newer GLM-5.3 and GLM-5.3-Flash entries, which are listed above GLM-5.2.

    AIWhy it matters: The page lists a dated series of Z.ai model releases, showing how the coding and long-horizon agent line has evolved from GLM-4.5 through GLM-5.2.

May 9

May 9Sat
  1. PaddlePaddleAI score60

    Baidu releases ERNIE 5.1 with reduced pretraining cost and parameter scale

    Baidu's PaddlePaddle account announced ERNIE 5.1, which it says cuts total parameters to about one-third and activated parameters to about one-half, using roughly 6% of the pretraining cost of models at similar scale. The post reports benchmark results including 99.6 on AIME26 with tools, surpassing DeepSeek-V4-Pro on τ3-bench and SpreadsheetBench-Verified, and ranking #4 globally on Arena Search. ERNIE 5.1 is available through the ERNIE website and Baidu AI Studio Model Playground.

Jan 13

Jan 13Tue

Jan 1

Jan 1Thu
  1. Moonshot AI (Kimi) · new models on Hugging FaceAI score75

    Moonshot AI releases open-source multimodal agent model Kimi K2.5

    Moonshot AI released Kimi K2.5, an open-source native multimodal agentic model built by continual pretraining on about 15 trillion mixed visual and text tokens. The model card reports a 1T-parameter Mixture-of-Experts architecture with 32B activated parameters and a 256K context length, and it lists benchmark results against GPT-5.2, Claude 4.5 Opus, Gemini 3 Pro, DeepSeek V3.2, and Qwen3-VL-235B-A22B-Thinking. Weights and code are released under a Modified MIT License, with API access on the Moonshot platform.

    AIWhy it matters: The model card gives a full benchmark table against GPT-5.2, Claude 4.5 Opus, and Gemini 3 Pro, useful for comparing open multimodal agent models.

Dec 14, 2025

Dec 14, 2025Sun
  1. FunAudioLLM (Alibaba Tongyi) · new models on Hugging FaceAI score38

    Alibaba Releases Fun-ASR-MLT-Nano-2512, an 800M Multilingual Speech Recognition Model

    Alibaba's FunAudioLLM released Fun-ASR-MLT-Nano-2512, an 800M-parameter multilingual speech recognition checkpoint on Hugging Face that supports 31 languages, with emphasis on East and Southeast Asian languages. It is trained on hundreds of thousands of hours of speech and is available through the FunASR toolkit. The source's benchmark tables cover the Fun-ASR family rather than this checkpoint, so no checkpoint-specific accuracy figures are reported.