SShogo Bench

DeepSeek V3 0324

DeepSeek chat model for instruction following, coding, and analysis

Shogo Agent Index

Unranked

Not ranked yet: 1 agentic benchmark measured, and we need at least 2 plus an overall capability score. It will be ranked as results are published.

Agentic coding
18
estimated
Long-horizon autonomy
28
50% measured
Knowledge-work agents
17
estimated
Computer use
7
estimated
Security & context learning
17
estimated

Price

Input
$0.31/M
Output
$1.21/M
Cache read
$0.15/M
3:1 blend
$0.53/M
Agent-effective
$0.22/M

Cheapest host for agents: GMICloud at $0.18/M.

Agent-effective = 20 input tokens per output token, 85% cache hits. Why.

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 50
Terminal-Bench 2.0est.8%predicted 9
DeepSWEest.6%predicted 10
FrontierSWEest.5%predicted 1
CursorBenchest.4%predicted 17
GSO-Benchest.4%predicted 2
METR Time Horizons49.6%10%source
Vending-Bench 2est.6%predicted 2
PostTrainBenchest.4%predicted 11
APEX-Agentsest.8%predicted 11
GDPvalest.7%predicted 17
Remote Labor Indexest.5%predicted 0
DeepResearch Benchest.5%predicted 41
OSWorld 2.0est.6%predicted 1
OSWorldest.4%predicted 16
Cybenchest.5%predicted 21
CL-benchest.5%predicted 13

Other benchmarks

BenchmarkCategoryScoreSource
Aider PolyglotCoding55.1%source
CritPtReasoning0.0%source
DTBenchKnowledge64.8%source
Fiction.LiveBenchLong context50.0%source
GPQA DiamondReasoning67.6%source
LMCAKnowledge15.5%source
MATH Level 5Math75.5%source
OTIS Mock AIMEMath37.8%source
SciCodeCoding39.0%source
SimpleBenchReasoning27.2%source

API hosts

Live prices and uptime for every host serving DeepSeek V3 0324 via OpenRouter.

HostInputOutputCache readAgent $/MUptime 24hQuant
GMICloud$0.29$1.14$0.11$0.18100.0%fp8
SiliconFlow$0.25$1.00—$0.2999.9%fp8