Matt Turck speaks with Mistral research lead Pierre Stock about Le Chonk, described as the West's largest open source model. Stock discusses the team training on 3,800 GPUs in its own data center, the fact that Le Chonk is still in training, what Mistral borrowed from DeepSeek, and the trade-off between open-sourcing weights and refusal behavior for cybersecurity. The episode also mentions that Mistral Large 5 will be released later.
The author reviewed the financial reports of the five major U.S. tech giants and found that more than 85% of Alphabet's and Amazon's quarterly profits come from unrealized gains on equity investments; their combined off-balance-sheet long-term purchase and lease commitments total approximately $1.65 trillion, already exceeding on-balance-sheet liabilities. Capital expenditure by the five giants is projected at $769.2 billion in 2026, while annual AI application revenue is about $200 billion, leaving a gap of roughly 5 times. After comparing China-U.S. pricing, the author notes that OpenAI charges $30 per million output Tokens while DeepSeek charges only $0.18, a price gap of 166 times, and concludes that the most likely endgame is that one node runs into trouble first and the problem then propagates through the chain of cross-shareholdings.
ities. Capital expenditure by the five giants is projected at $769.2 billion in 2026, while annual AI application revenue is about $200 billion, leaving a gap of roughly 5 times. After comparing China-U.S. pricing, the author notes that OpenAI charges $30 per million output Tokens while DeepSeek charges only $0.18, a price gap of 166 times, and concludes that the most likely endgame is that one node runs into trouble first and the problem then propagates through the chain of cross-shareholdings.
Taiwan's Newtalk published an article documenting that the author used a local Ollama to run Qwen3.5:4b to ask about the Cultural Revolution figure Wu Shufang; the model first said it could not confirm whether this person existed, and after being changed to a four-choice format, after hundreds of lines of reasoning it returned only a single letter. The author also submitted 200 questions to DeepSeek, Kimi, Qwen and GLM, and found that more than one-third refused to answer, used official wording, or gave falsifiable answers; on DeepSeek this proportion reached 63%. The Christian Science Monitor reported that Chinese models are spreading through developers and enterprises, and OpenRouter data shows that usage by US enterprises has grown nearly tenfold since early 2025.
person existed, and after being changed to a four-choice format, after hundreds of lines of reasoning it returned only a single letter. The author also submitted 200 questions to DeepSeek, Kimi, Qwen and GLM, and found that more than one-third refused to answer, used official wording, or gave falsifiable answers; on DeepSeek this proportion reached 63%. The Christian Science Monitor reported that Chinese models are spreading through developers and enterprises, and OpenRouter data shows that usa
A Huxiu analysis points out that domestic computing power can now cover large model operation: Zhipu's domestic-chip inference cluster has reached a scale of 100,000 cards, and Meituan's LongCat-2.0 has also completed training on domestic computing power. However, Ascend 910C achieves an MFU of about 30%-35% in trillion-parameter post-training, lower than H100's 40%-50%. The article breaks computing efficiency into three variables—chips, software, and models—and compares the four paths of Zhipu, MiniMax, DeepSeek, and Kimi, concluding that Zhipu has the highest short-term certainty, while DeepSeek has the greatest medium-term elasticity. The author proposes three tracking indicators—capacity, efficiency, and cost—and argues that only when all three improve simultaneously can domestic computing power shift from a backup option to a cost advantage.
r post-training, lower than H100's 40%-50%. The article breaks computing efficiency into three variables—chips, software, and models—and compares the four paths of Zhipu, MiniMax, DeepSeek, and Kimi, concluding that Zhipu has the highest short-term certainty, while DeepSeek has the greatest medium-term elasticity. The author proposes three tracking indicators—capacity, efficiency, and cost—and argues that only when all three improve simultaneously can domestic computing power shift from a backup
