NVIDIA、Anthropic、Google 和 Ideogram 发布新模型和研究

报道来源 [9]

1. Latent Space (swyx) TIER_1 English(EN) · 2026-06-06 04:34

   [AINews] 今日无甚大事发生

   a quiet day of RSI.

2. Latent Space (swyx) TIER_1 English(EN) · 2026-06-05 06:44

   [AINews] 今天没发生什么大事

   a quiet day

3. Smol AINews TIER_1 English(EN) · 2026-06-05 05:44

   今天没发生什么大事

   **Anthropic's Mythos/Opus cycle** sparked mixed reactions with praise for **Claude Mythos**'s one-shot workflows and concerns over **Opus 4.8** benchmark regressions. **Opus 4.7** showed strong chemistry task performance, "making Claude a chemist." **Sakana AI** launched an **RSI…

4. Smol AINews TIER_1 English(EN) · 2026-06-04 05:44

   今天没发生什么大事

   **NVIDIA** released **Nemotron 3 Ultra**, a fully open **550B MoE** model with **55B active parameters** and **1M context**, optimized for long-running agent tasks with up to **5x speedup** and **30% cost reduction**. It features hybrid Mamba/attention, LatentMoE, native MTP, and…

5. Smol AINews TIER_1 English(EN) · 2026-06-02 05:44

   今天没发生什么大事

   **Microsoft** released the detailed technical report for **MAI-Thinking-1**, a generalist reasoning model trained without third-party distillation, achieving **97% on AIME 2025** and outperforming Sonnet 4.6 in human preference tests. The report was praised for transparency, reve…

6. Smol AINews TIER_1 English(EN) · 2026-06-01 05:44

   今天没发生什么大事

   **NVIDIA** led open-source AI model releases with **Cosmos 3**, a comprehensive omnimodal world model unifying language, image, video, audio, and action using a Mixture-of-Transformers design, and **Nemotron 3 Ultra**, a **550B** parameter open-weight model noted for high serving…

7. Smol AINews TIER_1 English(EN) · 2026-05-29 05:44

   今天没发生什么大事

   **Anthropic** rolled out **Claude Opus 4.8**, which shows incremental improvements but mixed benchmark results, including better cooperation and coding behavior but some regressions in document parsing. Platform updates include mid-conversation system instructions enhancing long …

8. Smol AINews TIER_1 English(EN) · 2026-05-26 05:44

   今天没发生什么大事

   **Harness engineering** is emerging as the key differentiator for coding agents, emphasizing the stack of **model + harness + eval loop** over just stronger base models. **DeepSeek** is building a harness team to optimize interaction and verification loops, while **Google's Gemin…

9. Smol AINews TIER_1 English(EN) · 2026-05-26 05:44

   今天没发生什么大事

   **Inference optimization** is increasingly architectural, with **EAGLE 3.1** improving speculative decoding and long-context handling, collaborating with **vLLM** and **TorchSpec**. **Perplexity** open-sourced a rebuilt **Unigram tokenizer** cutting CPU use by **5–6×** and achiev…