SShogo Bench

GPT-5.4 vs Qwen3.8 Max Preview

Qwen3.8 Max Preview leads GPT-5.4 by 2.0 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-5.4 wins 8 and Qwen3.8 Max Preview wins 7 of 15 shared benchmarks. Qwen3.8 Max Preview is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-5.4Qwen3.8 Max Preview
Agentic coding5758
Long-horizon autonomy6163
Knowledge-work agents4046
Computer use4241
Security & context learning5452

Shared benchmarks

BenchmarkGPT-5.4Qwen3.8 Max Preview
APEX-Agentsindex52.4%63.3%
DeepSWEindex51.8%57.5%
Chess Puzzles48.0%46.0%
CritPt23.4%20.0%
DTBench94.4%92.0%
FrontierMath Tier 449.0%46.3%
FrontierMath Tiers 1-378.6%74.7%
GPQA Diamond93.3%92.7%
LMCA52.0%46.2%
MirrorCode23.3%52.2%
OTIS Mock AIME97.8%99.4%
ProofBench56.0%58.0%
SciCode56.6%53.2%
SimpleQA Verified45.1%45.8%
WebDev Arena14631671

Specs and price

GPT-5.4Qwen3.8 Max Preview
Agent Index51.6 (#33)53.6 (#30)
Agent $/M$1.27Free
Input / output $/M$2.50 / $15.00Free / Free
Context1.1M1M
ReleasedMar 5, 2026Jul 19, 2026
WeightsClosedClosed
API hosts41

Don’t pick one. Route.

Shogo can run GPT-5.4 and Qwen3.8 Max Preview side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo