CLM-8B: contrastive System One model claims 9x faster inference, agentic SOTA
ChengleiSi · x · 2026-09-24
A new Contrastive Language Model (CLM) uses a contrastive objective connecting states and actions.
- CLM-8B, pre-trained on internet-scale data, delivers up to 9x faster inference than Jev with comparable performance on computer-use, gaming, and tool-calling tasks
- With lightweight fine-tuning it sets SOTA on agentic coding benchmarks: DeepSWE (81.6%) and Terminal-Bench 2.1 (87.6%); Jev fails as an effective verifier for these long-horizon tasks
- Training/serving infra disaggregates states and actions so embeddings can be cached independently, cutting latency when state evolves but the action set is fixed
More from Models
- antirez backs low-bit DeepSeek 4.1 Flash over Qwen 27B dense on Apple M5 Macs — antirez · 2026-11-12
- Gemini re-renders images as text when vision is removed, dropping only 6% WSR — zacharynado · 2026-11-12
- Bare-harness coding benchmark: Fable 5.2 best overall, Gemini 4 Carbon best value — Semantic_meaning · 2026-11-12
- Google's Physical-Engineering's Last Exam: all 14 frontier models score below 50% — dl_weekly · 2026-11-12
- Reddit's 2026 LLM hall of fame: Opus 5.5 still unbeaten after 9 rival releases — G0dZylla · 2026-11-12
- Commit leak reveals unreleased Gemini 4 variants: Argon, Hydrogen and Helix (unconfirmed) — BlackHC · 2026-11-11