The open-weight race is accelerating on two fronts: compact models that train on single GPUs, and Chinese labs such as MiniCPM, DeepSeek, Qwen and Moonshot closing gaps in scale and revenue. This challenges Western-only model assumptions and reshapes on-premise deployment choices.
Moonshot AI reports more than US$1 billion in annual recurring revenue as of August and targets US$2 billion by end-2026. The gains come from Kimi subscriptions and services built on the open-weight K3 model. The licence imposes commercial agreements...
A 150-million-parameter model with performance similar to GPT-2 Small has been trained on a single RTX Pro 6000 using 7 billion tokens. Published benchmarks: PIQA 62.24%, Hellaswag 32.20%, Arc-Easy 44.91%, Arc-Challenge 25.00%, Arithmark 3.0 33.90%, ...
An independent researcher has prepared a dense 9.4-billion-parameter model for training on a single card, with distillation from Llama 3 and architectural simplifications. No benchmarks are public, but the signal for on-premise AI is clear: the bottl...
An independent researcher has opened a roughly 9.4B parameter dense model designed for single-card training. The project uses a Llama 3 tokenizer, distilled logit data, and architecture cues from Moonshot and Qwen. The real test is not technical; it ...
Teams running H100 clusters with policies that exclude Chinese models start at a disadvantage: in the open 120B+ vision segment, the gap with GLM, Qwen, and DeepSeek is measurable. Among the Western candidates cited, Inkling Small and Command A+ cove...
In the BabyLM 2026 Strict-Small, Qiushi Engine gained 0.23 Overall points across nine metrics without increasing scale, but by reorganizing experience around contextual dependencies. For teams evaluating self-hosted LLMs, the relevant signal is the d...
Qiushi Engine ran an end-to-end autonomous research program on BabyLM 2026 Strict-Small, using 10 million corpus words and 100 million cumulative word presentations. Three stages linked frontier advancement, principle discovery, and principle-guided ...
The page exists, the model is called DeepSeek-V4.1-Flash. The source provides no technical specifications: no VRAM, no context, no metrics. The 'Flash' label suggests an inference-oriented variant, but for on-premise evaluators it raises more questio...
OpenBMB has released MiniCPM5-2B, a 2-billion-parameter open weights model scoring 15 on the Artificial Analysis Intelligence Index v4.2, the highest among open models up to 4B. For self-hosted deployments, the small size and open weights lower the p...
