Western open-weights model Inkling 2S matches DeepSeek-level GPU kernel optimization
AccBalanced · x · 2026-11-07
Soumith Chintala demoed a model working for hours on GPU kernel optimization. Commenters note Inkling 2S, from one-year-old Thinky, achieves parity with DeepSeek's kernel optimization capability at similar scale and lower training cost — a major leap for Western open weights and for agentic long-horizon engineering tasks.
More from Models
- Antigravity's new UI has an annoying bug: images can't be removed from the chatbox — cyrus_zei · 2026-11-08
- Stories about a generic AI beat self-referential stories, challenging the Persona Selection Model — repligate · 2026-11-08
- X user ranks Chinese AI labs: Zhipu and DeepSeek in A tier, Qwen only C — xeophon · 2026-11-07
- Qoder lands stealth model 'Qoduck', free daily for individual users for limited time — hey_abusiddik · 2026-11-07
- Gemini Locks Chat Thread of Google AI Pro User at ~80k Tokens, Demands Upgrade — Beginning-Row-4125 · 2026-11-07
- Open-weight models close the cyber gap: DeepSeek V4.1 Flash hits 72.8% at a tenth the cost — The Decoder · 2026-11-07