Grok 4.5

xAI released 2026-07-08 proprietary

$3.00 per million tokens, balanced

Gemini 3.7 Flash is both better and cheaper.

It scores +0.2 higher and costs 75% less ($0.750/M against $3.00/M) on this workload — and it does everything this model does.

GLM 5.3 is cheaper still (28% less) but drops no image, file input.

Every price dimension

Input$2.00/M
Output$6.00/M
Cached input$0.300/M
Web search$0.005/call

Past 200,000 tokens the price changes. Input goes to $4.00/M (2×) and output to $12.00/M. The headline rate does not apply to a long-context workload.

Independent scores

Intelligence55.8
Coding72.4
Agentic48.9
LMArena Elo1452.3default

Sources are listed separately rather than averaged. Across the 62 models both have scored they correlate at r = 0.806 — close agreement overall, but four models rank very differently between them, and a blended score would hide exactly those.

Best rankings by task

  • godotgamedev #3of 43 1270
  • androidnative #5of 54 1266
  • agenticgamedev #6of 34 1220
  • htmlslides #6of 34 1216
  • asciiart #7of 83 1294
  • mobileapps #7of 56 1260
  • python-pptxslides #8of 35 1249
  • fullstack #9of 59 1277

Capability

Context window500K
Input modestext, image, file
Tool useyes
Reasoningalways on
Open weightsno

Provenance

Price sourceopenrouter.ai
Fetched2026-08-24
Quality dataverified
Cross-checkedvendor page

Grok 4.5 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM.

Markdown for LLMs