GPT-5

OpenAI released 2025-08-07 proprietary

$3.44 per million tokens, balanced

Grok 4.6 is both better and cheaper.

It scores +25.6 higher and costs 13% less ($3.00/M against $3.44/M) on this workload — and it does everything this model does.

Gemini 3.7 Flash is cheaper still (78% less) but drops 128K → 66K max output.

The same model via batch is 50% cheaper ($1.72/M) — same weights, different latency.

Every price dimension

Input$1.25/M
Output$10.00/M
Cached input$0.125/M
Web search$0.01/call
Batch discount50%

Independent scores

Intelligence35.3
Coding37.8
Agentic26.5
LMArena Elo1405.6high

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

  • svg #21of 98 1229

Capability

Context window400K
Max output128K
Input modestext, image, file
Tool useyes
Reasoningalways on
Knowledge cutoff2024-09-30
Open weightsno

Provenance

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

GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy...

Markdown for LLMs