Huawei pulls Ascend 960DT forward to Q1 2027 with 2x compute, but smaller SuperPoD
GlennLuk · x · 2026-09-17
Huawei's Ascend 960 roadmap update is being read as a bellwether for China's AI compute supply chain:
- 960DT pulled forward from Q4 2027 to Q1 2027 — three quarters early
- Substantial spec jump: 2x compute, memory bandwidth, memory capacity, and interconnect ports vs the 950
- Open question: last year Huawei said the Atlas 960 SuperPoD would scale to 15,488 Ascend 960s, but the announcement refers to a 4,096-chip SuperPoD — the chip arrives far earlier, yet the announced SuperPoD is much smaller than planned
- Commentators read the acceleration as signaling confidence in domestic HBM supply (and DUV scanner supply despite 2024 export controls)
Related event: Huawei Unveils Ascend 960 and UnifiedBus for Million-Chip Clusters(4 posts)→
More from Infra
- Scotland pauses new data centre approvals for a year; Australia urged to follow — TobyWalsh · 2026-09-17
- huggingface_hub 1.32 lets UV scripts declare runtime images for HF Jobs — vanstriendaniel · 2026-09-17
- Open-sourced Qwen-1B-RLCD runs type-safe JSON inference 5x faster on-device — JiliJeanlouis · 2026-09-17
- Fathom Speeds Million-Token KV Scans 1.67x with Per-Query Read Depth — Vivek Kalyanarangan · 2026-09-17
- Edge0 Streams a 35B MoE from SSD at 20 tok/s on a Single 24GB GPU, Open Source — Edge0 · 2026-09-17
- Daytona Benchmarks NVIDIA Vera CPU on Agentic Workloads, Cited in Official Blog — mattturck · 2026-09-17