COGNITION LAUNCHES SWE-1.7, CLAIMS FRONTIER PERFORMANCE AT LOWER COST

SWE-1.7 represents Cognition's most capable model to date, built on foundation work from Kimi K2.7. The model achieves what Cognition describes as "frontier-level intelligence" across benchmarks while maintaining lower operational expenses than comparable systems. The 1,000 tokens-per-second throughput rate addresses a key bottleneck in AI deployment, enabling faster inference for production applications. Cognition positions SWE-1.7 as competitive with leading closed-source models from OpenAI and Anthropic. The release continues Cognition's focus on building high-performance AI systems optimized for developer productivity. Pricing details remain limited, though the company emphasizes cost advantages as a core differentiator. The model's performance on standardized benchmarks will likely face independent verification as adoption expands.