Back

SpaceXSI

Grok 4.7

ReasoningClosed weights
Intelligence
42.2
Rank 51 of 744 measured models
Output speed
62.1 t/s
Rank 274 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
$1.25
Rank 143 of 193 measured models

Grok 4.7 (Low) scores 42.2 on the Artificial Analysis Intelligence Index, ranking 51st of 744 measured models. The closest models on this index are GPT-5.6 Terra (42.1), Claude Opus 4.8 (batch) (41.8), and GLM-5.3-Flash (41.8). At $3 per million tokens (3:1 input/output blended price), it is cheaper than GPT-5.6 Terra ($4.50), more expensive than GLM-5.3-Flash ($0.237). With an output speed of 62.1 t/s, it is about as fast as GLM-5.3-Flash (62.3 t/s), slower than GPT-5.6 Terra (107 t/s).

Artificial Analysis Indices
Low
Intelligence42.2
Rank 51 of 744 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Cost per task
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting48.0
Rank 36 of 218 measured models
Strategy & Ops50.8
Rank 39 of 218 measured models
Legal50.7
Rank 38 of 218 measured models
Engineering42.5
Rank 56 of 217 measured models
Economics52.8
Rank 42 of 225 measured models
Standard Benchmarks
HLE39.2%
Rank 90 of 664 measured models
SciCode54.9%
Rank 56 of 231 measured models
AA-LCR80.0%
Rank 77 of 575 measured models
Terminal-Bench v416.2%
Rank 67 of 218 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy