SpaceXSI
Grok 4.20 0309 v2
ReasoningClosed weights
- Intelligence
- 25.7
- Rank 167 of 744 measured models
- Output speed
- 101 t/s
- Rank 165 of 403 measured models
- Price/1M
- $1.56
- Rank 441 of 659 measured models
- Context window
- 2M
- Rank 8 of 1199 measured models
Grok 4.20 0309 v2 (Reasoning) scores 25.7 on the Artificial Analysis Intelligence Index, ranking 167th of 744 measured models. The closest models on this index are GPT-5.5 Instant (June 2026) (26.0), Kimi K2.7 Code (25.8), and Inkling Small (25.7). At $1.56 per million tokens (3:1 input/output blended price), it is cheaper than GPT-5.5 Instant (June 2026) ($11.25) and Kimi K2.7 Code ($1.71), more expensive than Inkling Small ($0.525). With an output speed of 101 t/s, it is faster than Kimi K2.7 Code (58.2 t/s), slower than GPT-5.5 Instant (June 2026) (151 t/s) and Inkling Small (190 t/s).
Additional benchmarks14
Design Arena · 3d52.1%
Design Arena · Androidnative35.0%
Design Arena · Asciiart48.2%
Design Arena · Codecategories51.8%
Design Arena · Dataviz52.0%
Design Arena · Fullstack41.1%
Design Arena · Gamedev50.1%
Design Arena · Godotgamedev38.5%
Design Arena · Htmlslides45.7%
Design Arena · Mobileapps45.4%
Design Arena · Svg52.4%
Design Arena · Uicomponent49.0%
Design Arena · Webapps49.7%
Design Arena · Website51.8%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- GPT-5.5 Instant (June 2026)
OpenAI · $11.25 · 151 t/s
26.0+0.3$11.25151 t/sCompare - MiMo-V2.5-Pro
Xiaomi · $0.544 · 38.8 t/s
26.0+0.3$0.54438.8 t/sCompare - Kimi K2.7 Code
Moonshot AI · $1.71 · 58.2 t/s
25.8+0.1$1.7158.2 t/sCompare - Inkling Small
Thinking Machines · $0.525 · 190 t/s
25.7$0.525190 t/sCompare
