Pass IndexThe State of AISign in

Grok 4.5

SpaceXAI's smartest model with frontier performance on coding, knowledge work, and STEM.

text + image + file → text · made by xAI

$2$6per Mtok in / out
xAI · 9 sellers

Sold by 12 ways

SellerLaneRate
xAI apistandard$2per Mtok in$6per Mtok out$0.3per Mtok cacheddocs.x.ai · read 2026-08-25
xAI apilong-context$4per Mtok in$12per Mtok out$0.6per Mtok cacheddocs.x.ai · read 2026-08-25
ElectronHub aggregatorstandard$2per Mtok in$6per Mtok outapi.electronhub.ai · read 2026-09-16
GitHub Copilot aggregatorstandard$2per Mtok in$6per Mtok out$0.5per Mtok cachedmodels.dev · read 2026-09-16
Inference.net aggregatorstandard$2per Mtok in$6per Mtok out$0.3per Mtok cachedapi.inference.net · read 2026-09-16
Nous Research aggregatorstandard$2per Mtok in$6per Mtok out$0.000002per image$0.3per Mtok cachedinference-api.nousresearch.com · read 2026-09-12
OpenRouter 3 ways$2per Mtok in$6per Mtok out$0.3per Mtok cached
OpenRouter aggregatorstandard$2per Mtok in$6per Mtok out$0.3per Mtok cachedopenrouter.ai · read 2026-08-24
OpenRouter aggregatorxai/priority$4per Mtok in$12per Mtok out$0.6per Mtok cachedopenrouter.ai · read 2026-08-25
OpenRouter aggregatorxai/zdr/priority$4per Mtok in$12per Mtok out$0.6per Mtok cachedopenrouter.ai · read 2026-08-25
Requesty aggregatorxai$2per Mtok in$6per Mtok out$0.5per Mtok cachedrouter.requesty.ai · read 2026-08-28
Vercel AI Gateway aggregatorstandard$2per Mtok in$6per Mtok out$0.3per Mtok cachedai-gateway.vercel.sh · read 2026-08-25
Venice AI aggregatorstandard$2.27per Mtok in$6.8per Mtok out$0.34per Mtok cachedmodels.dev · read 2026-09-16

Measured 58 standings

PlaceBoardMetricScore
2ndof 17SWE-rebench Leaderboard (default window 15/05/2026–01/07/2026, 111 problems)Resolved Rate (%) (high)63.8
3rdof 42Design Arena · godotgamedev (agents)Elo1270
3rdof 28τ³-Banking LeaderboardPass^147.9
4thof 23Gbaeval — Epoch AIOverall score0.654
4thof 17terminal-bench@2.1Accuracy (high)79.3
5thof 161Dtbench — Epoch AIAccuracy (high)96.53
8thof 34LMArena · SearchScore (Elo)1213
9thof 26Surface evolver bench — Epoch AIMean score (high)0.744
10thof 36Design Arena · htmlslides (agents)Elo1208
10thof 12Posttrainbench — Epoch AIAverage (%) (high)0.234
11thof 24Blueprint bench 2 — Epoch AIScore27.332
11thof 40Design Arena · agenticgamedev (agents)Elo1235
11thof 55Design Arena · androidnative (agents)Elo1247
11thof 64Design Arena · fullstack (agents)Elo1261
12thof 30Frontiercode — Epoch AIMain score0.424
13thof 313GPQA Diamondmean_score (high)93.434
14thof 102Simplebench — Epoch AIScore (AVG@5)0.7
16thof 63Design Arena · mobileapps (agents)Elo1244
18thof 43Design Arena · python-pptxslides (agents)Elo1230
20thof 65Apex agents — Epoch AIPass@1 score0.342
22ndof 112Ale bench — Epoch AIPerformance (high)1309.28
22ndof 121Webdev arena — Epoch AIArena Score1556.36
23rdof 290OTIS Mock AIME 2024-2025mean_score (high)97.778
24thof 92Design Arena · asciiart (models)Elo1269
24thof 62Design Arena · webapps (agents)Elo1231
24thof 152LMArena · VisionScore (Elo)1279
24thof 123Lmca — Epoch AIScore (high)45.2
26thof 80Simpleqa verified — Epoch AImean_score (high)48.3
27thof 39ARC-AGI-3 (Semi-Private)Score (%) (medium)0.32
27thof 156Design Arena · codecategories (models)Elo1293
27thof 152Design Arena · dataviz (models)Elo1286
27thof 109Design Arena · svg (models)Elo1248
27thof 162Design Arena · website (models)Elo1296
27thof 39Gdp pdf — Epoch AIGDP.pdf score (high)0.14
27thof 128LMArena · WebDevScore (Elo)1555
28thof 64Proofbench — Epoch AIAccuracy (high)31
29thof 184Artificial Analysis · Coding IndexIndex72.4
31stof 147Design Arena · 3d (models)Elo1288
31stof 155Design Arena · gamedev (models)Elo1288
32ndof 150Design Arena · uicomponent (models)Elo1294
32ndof 106Frontiermath tiers 1 3 v2 — Epoch AImean_score (high)57.193
33rdof 143Artificial Analysis · Agentic IndexIndex42.1
34thof 168Scicode — Epoch AIScore (high)54.051
34thof 60Vending bench 2 — Epoch AIScore3887.426
36thof 222Chess puzzles — Epoch AImean_score (high)36
38thof 62Frontiermath tier 4 v2 — Epoch AImean_score (high)24.39
40thof 302Artificial Analysis · Intelligence IndexArtificial Analysis Intelligence Index (high)39
42ndof 402LMArena · TextScore (Elo)1468
43rdof 69Deepswe — Epoch AIPass@1 (high)0.538
50thof 177Critpt — Epoch AIAccuracy (high)15.429
52ndof 205ARC-AGI-1 (Public Eval)Score (%) (high)94.75
64thof 230ARC-AGI-1 (Semi-Private)Score (%) (medium)87.17
64thof 216ARC-AGI-2 (Public Eval)Score (%) (high)58.19
73rdof 232ARC-AGI-2 (Semi-Private)Score (%) (high)52.64
74thof 244Arc agi — Epoch AIScore (medium)87.17
81stof 227Arc agi 2 — Epoch AIScore (high)52.64
81stof 170Weirdml — Epoch AIAccuracy46.43
161stof 656Epoch capabilities index — Epoch AIECI Score (high)153.94

About

Grok 4.5 — a text model from xAI, sold by 9 companies from $2 in and $6 out per million tokens, placed 5th of 161 on Dtbench — Epoch AI.

It takes text, images and files and returns text, with a context window of 500,000 tokens. It was published in July 2026. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning, code and agents. Nine companies sell it. The cheapest is $2 in and $6 out per million tokens at ElectronHub. Beside the standard rate there are faster and separately routed lanes. It has been measured on 58 boards, and stands best at 5th of 161 on Dtbench — Epoch AI.

Every current figure

Maker
xAI
Register
model
Takes
text + image + file
Returns
text
Context
500,000 tokens
Published
July 2026
Licence
not read
Sellers
9
Price
$2 in and $6 out per million tokens — ElectronHub
Boards
58
Best place
5th of 161 — Dtbench — Epoch AI

Known as 11 names

Grok 4.5 (high)Grok 4.5 (Low)Grok 4.5 (Medium)grok-4.5Grok 4.5 [high]Grok Grok 4.5 [high]Cursor CLI + Grok 4.5 (high)x-ai/grok-4.5xai/grok-4.5spacexai/grok-4.5grok-4.5-latest