Zyphra scales subquadratic attention ZHA to trillion-param open models with 9-60x long-context speedups
realmrfakename · x · 2026-11-18
Zyphra announced it has scaled its subquadratic attention method, ZHA, to leading open-source multi-trillion-parameter models including Nemotron, Qwen, Mimo, and GPT-OSS. The company claims ZHA preserves reasoning and agentic performance while delivering 9-60x long-context speedups at 10M tokens.
More from Infra
- Podcast episode 3: why local AI, and why you should keep your own data — heyneighbor · 2026-11-18
- SpaceX CFO lays out orbital compute timeline: first satellites in 2027, meaningful compute in space by 2029 — XFreeze · 2026-11-18
- Lambda Deploys NVIDIA Photonic CPO Switches: 10% Lower Latency, 15x Reliability — TheZachMueller · 2026-11-18
- NVIDIA releases 4B speculative decoding draft boosting Qwen3.8-27B acceptance by 46% — _akhaliq · 2026-11-18
- Perplexity launches Orchestrator API: 45+ models, auto-routing, unified billing — perplexity_ai · 2026-11-18
- Replit's agent swarm rebuilds an $83,700/yr observability tool as open-source DataCat — amasad · 2026-11-18