#Other

AIMaintainers and NVIDIA engineers improved multi-GPU performance in ggml, the low-level engine behind llama.cpp, yielding significant gains on RTX systems. The work also lays groundwork for hardware-agnostic tensor parallelism in ggml. Details are in a technical blog from NVIDIA RTX Spark.

AIMax Woolf reports that Hy3, an unfamiliar LLM, has unexpectedly topped OpenRouter's AI Model Rankings by a large margin. After examining the usage data, he says it only deepened his confusion about the result. The full analysis is in his new blog post on minimaxir.com.

AICognition has released Auto-Triage in Devin Automations, which lets Devin respond to Slack messages, Linear events, GitHub activity, schedules, and webhooks. Devin can investigate with connected observability tools and the codebase, then post a summary, tag an owner, or open a PR. Devin runs in network-sandboxed environments with added protections against prompt injection and data exfiltration, and a limited-time offer gives $200 in credits for a first automation.

Why it matters: The post shows how an agent handles alerts and bug reports from existing team channels, a practical pattern for teams weighing automated incident response.

AIChip Huyen congratulated Alec Radford, Luke Metz, and Soumith Chintala on winning an ICLR Test of Time award for work published at ICLR 2016. She highlighted that the team's paper was produced by researchers in their twenties without any PhDs among them, and expressed hope that the three collaborate again.

AIKarpathy describes using LLMs to compile raw source documents into a markdown wiki that he views in Obsidian, with the LLM writing and maintaining most of the wiki. He reports that at about 100 articles and 400K words, the LLM agent can answer complex questions directly from the wiki, and he also runs LLM health checks to find inconsistencies and gaps. He shares the underlying idea as an "idea file" that users can give to their own agents to build a customized version.

AIAndrej Karpathy describes using LLMs to compile raw research sources into a markdown wiki of about 100 articles and 400K words, viewed in Obsidian. He says an LLM agent answers complex questions against the wiki without RAG, with outputs filed back to enhance it. He also suggests the workflow could become a product rather than a collection of scripts.

AITri Dao announced Mamba-3, which he described as the most powerful linear sequence model to date, as hybrid architectures increasingly rely on strong linear models. The post cites Qwen, Kimi-Linear, and NVIDIA's Nemotron-3 Super as examples of this trend. According to co-author Albert Gu, Mamba-3 shows noticeable performance gains over Mamba-2 and Gated DeltaNet at all sizes while maintaining speed.

AIMistral announces Voxtral 2 with two transcription models: Voxtral Realtime, released under an Apache 2 license with latency configurable to sub-200 ms, and Voxtral Mini Transcribe 2, which adds speaker diarization, word-level timestamps, and context biasing. The models support 13 languages and are available through the Mistral API, which the post describes as one of the most cost-effective transcription APIs on the market. The attached chart shows word error rates on FLEURS across Italian, Spanish, English, German, Portuguese, French, Russian, Dutch, and Chinese at several latency settings.

AIKarpathy argues animal intelligence is only one point in a large space of possible minds, and LLMs arise from a fundamentally different optimization process. He contrasts survival-driven animal drives with LLM training shaped by imitation of human text, RL on task distributions, and user engagement metrics, which he says leaves LLMs jagged and prone to sycophancy. He calls LLMs humanity's first contact with non-animal intelligence and says people who build accurate internal models of them will reason about them better.