ainewsblitz.com

Breaking

Cognition's SWE-1.7 Runs on Cerebras at 1,000 Tokens per Second, Sharpening Speed Race in AI Coding Agents

Cognition's SWE-1.7 Runs on Cerebras at 1,000 Tokens per Second, Sharpening Speed Race in AI Coding Agents
  • Software Dev & Coding
  • AI Agents
  • Foundation Models

Cognition's newly released SWE-1.7 coding model, running through Cerebras hardware at roughly 1,000 tokens per second, completed a sample development task in 13 seconds versus 38.5 seconds for Z.ai's GLM-5.2 Max — a demonstration that reframes latency, not just accuracy, as a competitive frontier for AI software engineers.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,934 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year