SShogo Bench

GPT-5.6 Luna vs Qwen3.8 Max Preview

Qwen3.8 Max Preview leads GPT-5.6 Luna by 2.2 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-5.6 Luna wins 7 and Qwen3.8 Max Preview wins 8 of 15 shared benchmarks. Qwen3.8 Max Preview is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-5.6 LunaQwen3.8 Max Preview
Agentic coding5958
Long-horizon autonomy6163
Knowledge-work agents3946
Computer use4041
Security & context learning5152

Shared benchmarks

BenchmarkGPT-5.6 LunaQwen3.8 Max Preview
APEX-Agentsindex43.0%63.3%
DeepSWEindex67.2%57.5%
Chess Puzzles43.0%46.0%
CritPt20.6%20.0%
DTBench89.1%92.0%
FrontierMath Tier 461.0%46.3%
FrontierMath Tiers 1-382.1%74.7%
GPQA Diamond91.6%92.7%
LMCA48.5%46.2%
MirrorCode21.1%52.2%
OTIS Mock AIME98.3%99.4%
ProofBench60.0%58.0%
SciCode53.6%53.2%
SimpleQA Verified41.0%45.8%
WebDev Arena15191671

Specs and price

GPT-5.6 LunaQwen3.8 Max Preview
Agent Index51.4 (#34)53.6 (#30)
Agent $/M$0.10Free
Input / output $/M$0.20 / $1.20Free / Free
Context1.1M1M
ReleasedJul 9, 2026Jul 19, 2026
WeightsClosedClosed
API hosts51

Don’t pick one. Route.

Shogo can run GPT-5.6 Luna 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