LMSYS Doubles Muse Spark Open Speed to 1,109 Tokens per Second
SGLang's new megakernel for Muse Spark Open collapses 1,450 kernel launches into 97 per decode step, doubling throughput on Blackwell GPUs.
20 stories
SGLang's new megakernel for Muse Spark Open collapses 1,450 kernel launches into 97 per decode step, doubling throughput on Blackwell GPUs.
Google Labs rolled out Tabletop, an AI-powered shared workspace that spins up custom planning tools like itineraries, polls, and brackets on demand.
Unsloth's new Dynamic V3.1 GGUFs squeeze Qwen3.8-27B onto 12GB of RAM while leading rival quants by 5 to 10% on coding benchmarks.
Runway open sourced Praxis-1, a 27B mixture-of-experts diffusion transformer that turns video pretraining into real robot control under Apache 2.0.
Google DeepMind's new conditional DNA model predicts variant effects in individual cell types, doubling causal variant recall and adapting to new single-cell data without retraining.
Databricks reframes Genie from a chat assistant into an always-on worker that monitors systems, drafts work, and acts under governed guardrails.
DeepMind's AlphaProof Code used Gemini agents to generate 6M lines of Lean proofs, verifying Signal's cryptographic protocol down to machine code.
vLLM's updated encoder-prefill-decode disaggregation cuts P90 time-to-first-token by 44-47% on multimodal workloads through metadata-only requests and cross-encoder caching.
The Tokyo startup brings in a former NSA director and a veteran CIA station chief to steer its push into Japanese defense and cyber work.
Baidu's new 280B sparse MoE activates just 14B parameters per token, serves above 200 tokens per second, and ships open weights by November 25.
ACE Music v1 tops the Artificial Analysis vocal music leaderboard with an Elo of 1160, edging out Suno v6 and the entire Mureka and Lyria lineups.
Alibaba opens preview access to Qwen4-Max across its Token Plan, QwenWork, Qoder, and Qwen App, with a focus on coding and agent workflows.
Cursor Projects now lets entire teams and their coding agents share one workspace, with cross-teammate agent coordination to prevent conflicts.
Microsoft Research shows that rewriting fewer than 10 dormant neurons can turn a well-behaved LLM into a sleeper agent that runs attacker code on a hidden trigger.
PyTorch's ExecuTorch now ships a WebGPU backend, letting a single exported .pte file run LLMs and vision models directly on the browser's GPU.
Google's newest video model climbs the Artificial Analysis boards with a 31% price cut, 15-second clips, and a #1 finish in complex edits.
Parallel's Search API takes the top two spots on an independent third-party search benchmark, with the cheaper Fast tier beating its own Advanced tier.
Microsoft Research built a platform that fabricates entire fake companies, employees, emails, and meetings so Copilot can be tested without touching real customer data.
Cognition is wiring Anthropic's restricted Mythos 5.2 model into Devin's managed vulnerability scans and seeding adoption with over $100M in credits.
Google DeepMind and Artificial Analysis launch MKA, a 200-task benchmark grading agents on visually dense professional work that takes experts hours to complete.