iFlytek releases device-cloud Spark X2.6
On 22 October 2026 iFlytek released Spark X2.6, a device-cloud model series, at its developer festival in Beijing. The same event also launched a speech foundation model and a digital-employee platform.
Total 42 · Clear filters
Meta releases the open-weight Muse Spark Open
On 15 October 2026 Meta released Muse Spark Open, an open-weight, natively multimodal model with 816 billion total parameters, optimised to run on Apple's Mac Studio. Its weights were published on Hugging Face.
Zhipu releases GLM-5, a 744B open model
On 12 February 2026 Zhipu released GLM-5, a 744-billion-parameter open model under the MIT License. It uses DeepSeek Sparse Attention, and its pre-training data grew from 23 trillion to 28.5 trillion tokens.
iFlytek releases Spark X2
On 11 February 2026 iFlytek released Spark X2, a large model trained entirely on domestic computing power and aimed at education, healthcare, automotive and agent use cases.
Alibaba unveils its Zhenwu PPU chip
On 29 January 2026 Alibaba's chip unit T-Head listed the Zhenwu 810E, formally revealing a PPU accelerator the company had kept under wraps and completing its combination of Qwen models, cloud and chips.
Unitree releases the H2 bionic humanoid
On 20 October 2025 Unitree announced the H2, a bionic humanoid robot with a face, 31 joints, a 180 cm frame and a 70 kg body, shown in a video performing ballet, punches, side kicks and a runway walk.
DeepSeek debuts sparse attention in DeepSeek-V3.2-Exp and halves API prices
DeepSeek released the experimental model DeepSeek-V3.2-Exp on 29 September 2025. It introduced DeepSeek Sparse Attention (DSA), a method aimed at cutting the cost of processing long inputs, and lowered API prices by more than 50 percent.
Huawei publishes three-year Ascend roadmap
At Huawei Connect 2025 in Shanghai, Huawei published its first three-year Ascend chip roadmap, running to 2028, alongside new Atlas SuperPoD systems and a pledge to open source CANN.
Zhipu releases GLM-4.6 with a 200K context and domestic-chip support
In late September 2025 Zhipu released GLM-4.6 under the MIT License, with a 200K-token context window, stronger coding and agent abilities, and support for Chinese domestic chips.
StepFun releases the open Step 3 multimodal reasoning model
In July 2025, during WAIC, StepFun released the open Step 3 multimodal reasoning model and formed a Model-Chip Ecosystem Innovation Alliance with Huawei, Biren, Moore Threads and Enflame.
NVIDIA announces Blackwell Ultra
NVIDIA announced Blackwell Ultra at GTC in San Jose on 18 March 2025, a platform aimed at AI training and at inference that spends more compute per answer, alongside new HGX B300 NVL16 systems and named server and cloud partners.
Huawei launches the Atlas 900 A3 SuperPoD
Huawei's Atlas 900 A3 SuperPoD, launched in March 2025, links up to 384 Ascend 910C chips into one logical computer rated at up to 300 PFLOPS, and is sold in cloud form as CloudMatrix384.
iFlytek releases Spark X1, a reasoning model trained on domestic compute
On 15 January 2025 iFlytek released Spark X1, described by the company as China's first deep reasoning model trained on fully domestic compute, built with Huawei and used first in education and healthcare.
Mistral AI releases Ministral 3B and 8B for edge devices
On 16 October 2024 Mistral AI released Ministral 3B and 8B, small models built for on-device and edge use, with 128K-token context, low per-token API prices and research access to the 8B Instruct weights.
Mistral AI and NVIDIA release Mistral NeMo 12B
On 18 July 2024 Mistral AI released Mistral NeMo, a 12B language model built with NVIDIA, with a 128K-token context window, a new multilingual tokenizer called Tekken, FP8-aware training and Apache 2.0 weights.
SoftBank acquires Graphcore
In July 2024 SoftBank Group acquired Graphcore, which became a wholly owned subsidiary and kept its name and offices; the deal was recorded at about US$500 million, subject to UK government review.
iFlytek releases Spark V4.0
On 27 June 2024 iFlytek released Spark V4.0, a large model it said was trained on a domestic 10,000-card cluster and beat GPT-4 Turbo on most measures while trailing on coding and multimodal work.
Unitree introduces the G1 humanoid robot
On 13 May 2024, Hangzhou-based Unitree Robotics launched the G1 humanoid robot at a starting price of 99,000 yuan, well below the 150,000–200,000 yuan that China Daily reported for many competing machines.
NVIDIA announces the Blackwell platform
On 18 March 2024 NVIDIA announced Blackwell, the GPU platform succeeding Hopper, claiming it can run real-time generative AI on trillion-parameter models at up to 25 times lower cost and energy.
NVIDIA announces Project GR00T
In March 2024 NVIDIA announced Project GR00T, a general-purpose foundation model for humanoid robots. A year later it released Isaac GR00T N1, which it presented as a fully customizable humanoid model.
iFlytek releases Spark V3.5, trained on domestic compute
On 30 January 2024 iFlytek released Spark V3.5, which it described as the first Spark model trained on the fully domestic FlyingStar One platform and which it compared with GPT-4 Turbo across several benchmarks.