modelbenchmark.io

Grok 4.6

xAI · grok-4-6

Compare

xAI's frontier model for long-running agents, coding, knowledge work, and visual projects

Specification

xai listing

Context
500K
Max output
500K
Released
2026-08-12
Knowledge cutoff
2026-02-01
Retires
—
Open weights
no
Input
text, image, pdf
Output
text

Price

US dollars per million tokens · xai list price

Input
$2.00
Output
$6.00
Cache read
$0.50
Cache write
$2.00

Batch and priority prices

US dollars per million tokens · per listing

PriceListing$/M
Priority inputbedrock_mantle · bedrock-mantle$3.85
Priority inputbedrock_converse · global$3.50
Priority outputbedrock_mantle · bedrock-mantle$11.55
Priority outputbedrock_converse · global$10.50

Long-context price tiers

the rate above each context threshold

AbovePrice$/M
200KCache read$1.00
200KInput$4.00
200KOutput$12.00

Quality

10 benchmarks · 2 sources

Composite
79th
percentile of 344 scored models
Rank
74 / 344
±0.07 sd
Evidence
10 × 2
3 or more benchmarks, from 2 or more sources
Effort range
0.6
points between effort settings
reasoning79th4/4 bench
math81st3/7 bench
knowledge66th1/1 bench
general76th1/1 bench

By reasoning effort

every setting placed on the same scale as the leaderboard

SettingCompositeBenchmarks behind it
high88th4
xhigh78th9

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
GPQA diamond94.0 ±1.4highEpoch AI2026-08-12
↳93.2 ±1.5xhighEpoch AI2026-08-14
Settings differ by 0.8 points on this benchmark.
OTIS Mock AIME 2024-202599.2 ±0.5xhighEpoch AI2026-08-14
↳97.8 ±1.0highEpoch AI2026-08-12
Settings differ by 1.4 points on this benchmark.
LiveBench79.0default · 23 runsLiveBench2026-06-25
FrontierMath-Tiers-1-3-v266.0 ±2.8xhighEpoch AI2026-08-14
Mystery Game Puzzles34.0 ±4.8xhighEpoch AI2026-08-14
SimpleQA Verified49.3 ±1.6highEpoch AI2026-08-27
↳48.9 ±1.6xhighEpoch AI2026-08-27
Settings differ by 0.4 points on this benchmark.
Chess Puzzles33.0 ±4.7highEpoch AI2026-10-13
↳33.0 ±4.7xhighEpoch AI2026-10-13
EBR-bench30.5 ±1.6xhighEpoch AI2026-08-18
Furniture Assembly40.0 ±6.3xhighEpoch AI2026-09-24
FrontierMath-Tier-4-v231.7 ±7.4xhighEpoch AI2026-08-14

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

Available from 48 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
xaivendor$2.00$6.00$0.50—500K500K—
302ai$2.00$6.00$0.50—500K500K—
abacus$2.00$6.00$0.50—500K33K—
aihubmix$2.00$6.00$0.50—500K500K—
aihubmix · aihubmix$2.00$6.00$0.50—500K500K—
amazon-bedrock · global$2.00$6.00$0.50—500K500K—
azure$2.00$6.00$0.50—200K128K—
azure_ai$2.00$6.00$0.50—200K128K—
azure-cognitive-services$2.00$6.00$0.50—200K128K—
bedrock_converse · global$2.00$6.00$0.50—500K500K—
cloudflare-ai-gateway$2.00$6.00$0.50—500K500K—
commandcode$2.00$6.00$0.50—500K500K—
crossmodel · x-ai$2.00$6.00$0.50$2.00500K500K—
edenai$2.00$6.00$0.50—500K500K—
github_copilot · github-copilot$2.00$6.00$0.50—500K500K—
github-copilot$2.00$6.00$0.50—500K128K—
google-vertex$2.00$6.00$0.50—524K500K—
kilo · x-ai$2.00$6.00$0.50—500K450K—
llmgateway$2.00$6.00$0.50—500K500K—
llmgateway-providers$2.00$6.00$0.50—500K500K—
llmgateway-providers · aws-bedrock$2.00$6.00$0.50—500K500K—
llmgateway-providers · vertex-openai$2.00$6.00$0.50—500K500K—
merge-gateway$2.00$6.00$0.50—500K500K—
nano-gpt · x-ai$2.00$6.00$0.50—500K450K—
nearai · x-ai$2.00$6.00$0.50—500K500K—
neon$2.00$6.00$0.50—500K524K—
oci$2.00$6.00$0.50—500K500K—
ofox · x-ai$2.00$6.00$0.50—500K66K—
opencode$2.00$6.00$0.50—500K500K—
openrouter · x-ai$2.00$6.00$0.50—500K450K—
opper$2.00$6.00$0.50—500K500K—
orcarouter · grok$2.00$6.00$0.50—500K500K—
perplexity$2.00$6.00$0.50————
perplexity-agent$2.00$6.00$0.50—500K500K—
prime · x-ai$2.00$6.00——500K450K—
requesty$2.00$6.00$0.50$2.00500K500K—
tempr$2.00$6.00$0.50—500K500K—
vercel · spacexai$2.00$6.00$0.50—500K500K—
vertex_ai$2.00$6.00$0.50—524K524K—
zenmux · x-ai$2.00$6.00$0.50—500K500K—
amazon-bedrock$2.20$6.60$0.55—500K500K—
amazon-bedrock · us$2.20$6.60$0.55—500K500K—
bedrock_converse · us$2.20$6.60$0.55—500K500K—
bedrock_mantle · bedrock-mantle
$2.20
$2.64
$6.60
$7.92
$0.55
$0.66
—500K500K—
venice$2.27$6.80$0.57—500K200K—
kenari$2.2857$6.8571$0.3429—500K500K—
databricks$2.50$7.50$0.625$2.50500K——
kyma · kyma$2.758$8.274——500K8K—

Price history

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

DateHostField$/M
2026-09-02bedrock_mantle · bedrock-mantleCache read$0.55
2026-09-09bedrock_mantle · bedrock-mantleCache read$0.55
2026-09-02bedrock_mantle · bedrock-mantleInput$2.20
2026-09-09bedrock_mantle · bedrock-mantleInput$2.20
2026-09-02kenariInputfree
2026-10-22kenariInput$2.2857
2026-09-02bedrock_mantle · bedrock-mantleOutput$6.60
2026-09-09bedrock_mantle · bedrock-mantleOutput$6.60
2026-09-02kenariOutputfree
2026-10-22kenariOutput$6.8571
2026-09-06azure_aiInput$2.00
2026-09-28azure_aiInput$1.25
2026-10-07azure_aiInput$2.00
2026-10-07azureInput$1.25
2026-10-08azureInput$2.00

More from xAI

most-hosted first