modelbenchmark.io

Grok-4

xAI · grok-4

Compare

Grok model for agentic tool use, reasoning, coding, and live assistance

Specification

most-agreed values

Context
256K
Max output
128K
Released
2025-07-10
Knowledge cutoff
2024-07
Retires
2027-01-31
Open weights
no
Input
text, image
Output
text

Price

US dollars per million tokens · most-agreed

Input
$3.00
Output
$15.00
Cache read
$0.75
Cache write
$15.00

Quality

5 benchmarks · 2 sources

Composite
82nd
percentile of 344 scored models
Rank
63 / 344
±0.07 sd
Evidence
5 × 2
3 or more benchmarks, from 2 or more sources
Effort range
—
one configuration only
coding98th1/3 bench
reasoning90th1/4 bench
math62nd3/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
Aider polyglot79.6high · aider diffAider2025-07-11
GPQA diamond87.0 ±2.0defaultEpoch AI—
OTIS Mock AIME 2024-202584.0 ±5.0defaultEpoch AI—
FrontierMath-v119.7 ±2.3defaultEpoch AI2025-11-13
FrontierMath-Tier-4-2025-07-011.0 ±0.0default · 2 runsEpoch AI2025-08-11

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 11 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
jiekou$2.85$14.25——256K8K—
abacus$3.00$15.00——256K16K—
azure_ai$3.00$15.00——262K8K2027-01-31
fastrouter · x-ai$3.00$15.00$0.75$15.00256K64K—
helicone$3.00$15.00$0.75—256K256K—
llmgateway$3.00$15.00$0.75—256K256K—
llmgateway-providers$3.00$15.00$0.75—256K256K—
oci · oci$3.00$15.00——128K128K—
poe$3.00$15.00$0.75—256K128K—
vercel_ai_gateway · vercel-ai-gateway$3.00$15.00——256K256K—
replicate · replicate$7.20$36.00—————

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02jiekouInput$2.70
2026-10-22jiekouInput$2.85
2026-09-02jiekouOutput$13.50
2026-10-22jiekouOutput$14.25

More from xAI

most-hosted first