Fireworks Launches Specialized Intelligence Index; DFS Model Hits 62.2% Vuln Detection Recall

nicolechirps · x · 2026-09-23

Fireworks AI launched the Specialized Intelligence Index (SII), a hub of real-work benchmarks across industries written by practitioner teams, comparing open, closed, and specialized models on quality, cost, and task duration.

DepthFirst Labs' flagship dfs-large1 (built on GLM 5.2 with RL post-training in partnership with Fireworks) posts 62.2% vulnerability detection recall at $6.77 per task and 75.6% differential analysis macro recall at $1.34 per task, claiming to push the performance-cost frontier on both tasks.

Related event: Fireworks AI Launches Specialized Intelligence Index for Real-World Benchmarks(5 posts)→

Original post →

More from Models

Models channel →