Skip to content

Benchmarks

Sources
8
Entries tracked
22,308
Last sync
19 min
All sources syncing normally
  • Arena · Text12 hours ago
  • Arena · WebDev12 hours ago
  • Arena · Vision12 hours ago
  • Arena · Agent12 hours ago
  • GPQA Diamond19 minutes ago
  • SWE-bench19 minutes ago
  • Terminal-Bench19 minutes ago
  • ARC-AGI-219 minutes ago
IFBench (Artificial Analysis) ×17 of 324 models
ModelOrg± CISwingAdd to compare
89DeepSeek V3.2 Specialedeepseek——63.9%——
99DeepSeek V3.2 (Reasoning)deepseekDec 2025$0.28 / $0.4260.7%—#86–247
112DeepSeek V3.1 Terminus (Reasoning)deepseekSep 2025$0.27 / $157.0%—#55–168
124DeepSeek V3.2 Exp (Reasoning)deepseekSep 2025$0.27 / $0.4154.1%—#92–334
181DeepSeek V3.1 (Reasoning)deepseekAug 2025—41.5%—#101–158
188DeepSeek V3 0324deepseekMar 2025—41.0%—#137–401
232DeepSeek V3 (Dec '24)deepseekDec 2024—34.8%—#178–644

Showing IFBench (Artificial Analysis) exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.

The few things in AI and quantum that actually mattered this week — honestly summarised.

One email a week. Unsubscribe in one click. No tracking, no sharing.