#DeepSeek

“A single cluster currently serves 155 trillion tokens/day, generating $38 in revenue per GPU-hour (at off-peak pricing).” Huge flex from DeepSeek. Not only did they open-source their inference engine, but they also open-sourced their revenue numbers. They didn't disclose which GPUs they're using here, but DeepInfer supports both NVIDIA Blackwell and Huawei Ascend. Even assuming they're renting B300s at $7 per GPU-hour, that's ~82% gross margin.

🚀 DeepSeek-V4.1-Flash-1017 来了！ 与上一版本相比，基准测试表现更强、任务完成更快、输出更短——在性能与成本上实现了帕累托改进。现已在我们的网站、应用和 API 上线。 🛠️ 针对 @DeepSeekHarness、Tencent CodeBuddy、WorkBuddy 等进行了优化。升级到 DeepSeek Harness v0.2.1-rc.1 或更高版本，以获得新模型的最佳体验。 1/4

Over the weekend I had Codex parse 500K arXiv AI/ML papers since ChatGPT to understand which open models are used for research. In 2024, ~30% of papers mentioned an American open model and only 10% a Chinese model. Today, ~40% of papers mention a Chinese (open) LLM, and only 25-30% an American one. Chinese models are the default for research. Chinese mentions are still growing while American open models are stagnating. When looking at this data it's important to remember that papers substantially lag model releases, as research takes a long time. Qwen's steady growth is reflective of this, but so is Llama's lasting power. Some more observations: 1. Qwen has been steadily growing, and today 1/3 of papers which mention any LLM mention qwen. OpenAI's closed models are the highest overall, at ~37%. 2. Llama peaked around April of 2025 at 30% of papers which mention any LLM (including ChatGPT etc). Llama 4 was released at about the same time, and Llama has been declining since. 3. Gemini and Claude are less common than the leading open models, mentioned in 10-15% of papers puts them behind all of Qwen, Llama, and DeepSeek. Open models should be and are the foundations of open research. The % of papers mentioning any LLM have been steadily climbing since 2023. | Year | January | April | July | October | | 2023 | 10.43% | 15.39% | 18.69% | 32.18% | | 2024 | 29.70% | 33.93% | 35.70% | 44.25% | | 2025 | 39.23% | 45.28% | 44.94% | 53.52% | | 2026 | 55.49% | 57.26% | 53.14% | TBD Now over 50% of AI papers, from 10% in 2023. Other notes: - Gemma and Mistral hover around 5-10%. - Our beloved fully-open Olmo models have been ~1% since the first release in Jan. 2024. - DeepSeek has a clear jump after R1 in Jan. 2025 - Data derived from the most popular ML arXiv categories: cs. AI, cs. CL, cs. CV, cs. LG, stat. ML Just like our downloads and derivative model data, this is updated daily on the Interconnects Open Model Dashboard.

https://x.com/i/article/2088681782606536704

Big news: DeepSeek-V4-Pro (Max) by @deepseek_ai is coming in around ~#8 overall (#2 among open models) in the Code Arena: WebDev! At 1607 pts, this places it after GPT-5.6 Sol (xHigh)(1622 pts), and makes it the second best open model after Kimi K3 (Max) (1674 pts). Note: this is an early AutoEval score, in which a Reward Model trained on Arena's human preference data casts automatic votes in place of live votes. We’ll continue to see how scores converge as more live human votes come in. See thread for Text Arena scores and more on AutoEval's methodology. Congrats to the @deepseek_ai team on this release!

Was told it takes only about 30K GPUs to pre train a mega model. Inference compute is the real monster. But then most likely they’ll just serve distilled smaller models.

DeepSeek V4 Flash from @deepseek_ai on ARC-AGI (Verified): - ARC-AGI-2: 61.4%, $0.04/task - ARC-AGI-1: 89.0%, $0.02/task DeepSeek V4 Flash sets the new standard on the cost-to-performance Pareto frontier.

注意到 V4 预览版论文里的 Table 6，V4 Pro (Preview) 和 Opus-4.6 的 SWE Verified 跑分， 80.6 比 80.8 正好差 0.3%。 这也不能说有些做中介的人写一句什么“性能跻身全球第一梯队，编程能力仅弱于 Claude 旗舰0.3%”是在纯瞎编是吧？

Open models have topped 2 new task leaderboards by real spend share: • Shell execution: DeepSeek V4 Pro • Tool dispatch: Kimi K3