Frontier Model Capability Premium Hits 100x Cost
ArtificialAnlys · x · 2026-07-07
Artificial Analysis released a capability index cost comparison: DeepSeek V4 Flash (max) completes tasks across six indices for under $0.04 each, yielding mid-tier scores, while GLM-5.2 (max) leads open-weight models at $0.26–$0.58 per task. On the "Strategy and Operations Index," Claude Fable 5 (falling back to Opus 4.8) outscored DeepSeek V4 Pro (max) by 12 points, but costs $3.48 per task compared to $0.03—a premium of over 100x, illustrating the massive cost premium of frontier capabilities.
Related event: Artificial Analysis Releases Six Industry Capability Indices(3 posts)→
More from Models
- Looking for a classifier of software engineering task shapes to pick models per task — StewartalsopIII · 2026-09-11
- DeepSeek V4 Pro API to continue after Sept 2026, billing unchanged — teortaxesTex · 2026-09-11
- DeepSeek V4.1 Flash Hits 98% of GPT-6 Astra's Score at 1.4% of the Cost in Third-Party Benchmark — ayushtweetshere · 2026-09-11
- TheZvi Polls: Has Your Coding Model Choice Changed Since Fable 5.1 and Astra? — TheZvi · 2026-09-11
- antirez Weighs In on Anthropic Banning Minors From Using Claude — antirez · 2026-09-11
- Engram's random reads don't suit SSDs; CPU-memory over NVLink could serve all 72 GPUs — bookwormengr · 2026-09-11