SShogo Bench

DeepSeek V4.1 Flash 1017 vs Gemini 3.8 Flash

DeepSeek V4.1 Flash 1017 leads Gemini 3.8 Flash by 1.4 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. DeepSeek V4.1 Flash 1017 wins 7 and Gemini 3.8 Flash wins 6 of 13 shared benchmarks. DeepSeek V4.1 Flash 1017 is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarDeepSeek V4.1 Flash 1017Gemini 3.8 Flash
Agentic coding6363
Long-horizon autonomy6664
Knowledge-work agents4947
Computer use4544
Security & context learning5453

Shared benchmarks

BenchmarkDeepSeek V4.1 Flash 1017Gemini 3.8 Flash
APEX-Agentsindex69.8%64.3%
CursorBenchindex47.7%39.6%
ARC-AGI94.5%98.5%
ARC-AGI-278.1%89.2%
Chess Puzzles46.0%58.0%
CritPt24.0%18.3%
FrontierMath Tier 446.3%22.0%
FrontierMath Tiers 1-371.2%68.4%
GPQA Diamond93.2%95.4%
OTIS Mock AIME99.4%98.9%
SciCode55.9%56.6%
SimpleQA Verified42.5%69.7%
WeirdML v39.1%7.4%

Specs and price

DeepSeek V4.1 Flash 1017Gemini 3.8 Flash
Agent Index57.6 (#17)56.2 (#21)
Agent $/M$0.050$0.35
Input / output $/M$0.15 / $0.60$0.75 / $3.75
Context1M1M
ReleasedOct 17, 2026Sep 2, 2026
WeightsOpenClosed
API hosts516

Don’t pick one. Route.

Shogo can run DeepSeek V4.1 Flash 1017 and Gemini 3.8 Flash side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo