SShogo Bench

Kimi K3 vs Qwen3.8 Max Preview

Kimi K3 leads Qwen3.8 Max Preview by 1.3 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Kimi K3 wins 10 and Qwen3.8 Max Preview wins 6 of 16 shared benchmarks. Qwen3.8 Max Preview is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarKimi K3Qwen3.8 Max Preview
Agentic coding6358
Long-horizon autonomy6563
Knowledge-work agents4246
Computer use4341
Security & context learning5352

Shared benchmarks

BenchmarkKimi K3Qwen3.8 Max Preview
APEX-Agentsindex47.8%63.3%
DeepSWEindex68.5%57.5%
FrontierSWEindex25.9%15.8%
Chess Puzzles37.0%46.0%
CritPt23.4%20.0%
DTBench91.2%92.0%
FrontierMath Tier 439.0%46.3%
FrontierMath Tiers 1-372.2%74.7%
GPQA Diamond93.1%92.7%
LMCA52.7%46.2%
MirrorCode54.4%52.2%
OTIS Mock AIME97.2%99.4%
ProofBench87.0%58.0%
SciCode59.5%53.2%
SimpleQA Verified50.6%45.8%
WebDev Arena16741671

Specs and price

Kimi K3Qwen3.8 Max Preview
Agent Index54.9 (#26)53.6 (#30)
Agent $/M$1.39Free
Input / output $/M$3.00 / $15.00Free / Free
Context1M1M
ReleasedJul 16, 2026Jul 19, 2026
WeightsOpenClosed
API hosts291

Don’t pick one. Route.

Shogo can run Kimi K3 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