Inference Index
Compare

GPT-5 Turbo vs Qwen 3 Max

OpenAI vs Qwen. Specs, benchmarks, and real per-task cost — all in one page.

Verdict

GPT-5 Turbo leads on LMSYS ELO (1398 vs 1295). Qwen 3 Max is ~3.7× cheaper on a 3:1 input:output blend. Qwen 3 Max is open-weights, the other is proprietary.

OpenAI
GPT-5 Turbo

OpenAI’s flagship unified model. Handles text, vision, and audio natively. The generalist benchmark champion.

Native multimodalStrong math and scienceHuge ecosystem
Qwen
Qwen 3 Max

Alibaba’s flagship. Strongest non-English coverage of any open model; dominant in Asian markets.

MultilingualOpen weightsCompetitive coding
Pricing
Input / 1M$6.00$1.60
Output / 1M$24.00$6.40
Context400K128K
Max output32K16K
LicenseProprietaryOpen weights
Released2026-01-212025-11-20
Benchmarks
LMSYS ELO1398.01295.0
MMLU Pro91.284.0
HumanEval93.087.2
SWE-Bench65.8
MATH91.585.1
GPQA68.1
MMMU78.871.8
IFEval88.985.5
Per-task cost
Summarize a 1-hour meeting transcript$0.102$0.027
Review a 500-line pull request$0.096$0.026
Answer a customer support ticket$0.037$0.0099
Extract structured data from a resume$0.054$0.014
Debug a stack trace with context$0.084$0.022

Per-call cost using published token counts for each task. Real-world prompts vary.