SpaceXSI
Grok 4.6
ReasoningClosed weights
- Intelligence
- 42.8
- Rank 47 of 744 measured models
- Coding
- 74.4
- Rank 31 of 298 measured models
- Output speed
- 52 t/s
- Rank 309 of 403 measured models
- Price/1M
- $3
- Rank 487 of 659 measured models
- Context window
- 500K
- Rank 349 of 1199 measured models
- Cost per task
- $1.14
- Rank 138 of 193 measured models
Grok 4.6 (Medium) scores 42.8 on the Artificial Analysis Intelligence Index, ranking 47th of 744 measured models. The closest models on this index are Kimi K3 (43.6), Claude Haiku 5.5 (43.4), and GPT-5.6 Terra (42.1). At $3 per million tokens (3:1 input/output blended price), it is cheaper than Kimi K3 ($6) and GPT-5.6 Terra ($4.50), more expensive than Claude Haiku 5.5 ($0.20). With an output speed of 52 t/s, it is faster than Kimi K3 (43.3 t/s), slower than Claude Haiku 5.5 (236 t/s) and GPT-5.6 Terra (107 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Step 5 Preview
StepFun · $1.43 · 89.3 t/s
43.7+0.9$1.4389.3 t/sCompare - Kimi K3
Moonshot AI · $6 · 43.3 t/s
43.6+0.8$643.3 t/sCompare - Claude Haiku 5.5
Anthropic · $0.20 · 236 t/s
43.4+0.6$0.20236 t/sCompare - GPT-5.6 Terra
OpenAI · $4.50 · 107 t/s
42.1−0.7$4.50107 t/sCompare
