Grok 4.20 Non-Reasoning
xAI Grok 4.20 non-reasoning model via Vertex AI (global endpoint)
VisionToolsStructured output
Context window
2M
Max output
30K
Input price
$2.00 /1M
Output price
$6.00 /1M
Benchmarks
Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.
Intelligence Index29.0
Coding Index22.0
GPQA Diamond
78%
Humanity's Last Exam
24%
SciCode
33%
IFBench
49%
Long-context reasoning
17%
Terminal-Bench Hard
17%
τ²-Bench Telecom
60%
See full leaderboard →Benchmarks via Artificial Analysis · View on AA
Available routes
The same Grok 4.20 Non-Reasoning runs on 3 different routes through the Opper gateway. Each route has its own data-handling posture.
| Provider | Region | Zero data retention | Training | Logging | Contract | Input | Output |
|---|---|---|---|---|---|---|---|
| Google Cloud | EU | Opt-in | No | Abuse-monitoring logs (window not stated) + 24h in-memory cache. Search/Maps grounding adds 30-day retention with no opt-out. | PAYG | $2.00 | $6.00 |
| xAI | US | Opt-in | No (opt-in) | Up to 30 days for abuse monitoring | Enterprise | $2.00 | $6.00 |
| xAI | EU | Opt-in | No (opt-in) | Up to 30 days for abuse monitoring | Enterprise | $2.00 | $6.00 |
Human review: Abuse-flagged only