SShogo Bench

Kimi K3

Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work

Shogo Agent Index

Supported#26 of 92
54.9range 47.3–62.5 · 29% of weight measured
Agentic coding
63
31% measured
Long-horizon autonomy
65
50% measured
Knowledge-work agents
42
32% measured
Computer use
43
estimated
Security & context learning
53
estimated

Price

Input
$3.00/M
Output
$15.00/M
Cache read
$0.30/M
3:1 blend
$6.00/M
Agent-effective
$1.39/M

Cheapest host for agents: Wafer at $0.60/M.

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

Agentic benchmarks

BenchmarkScoreNormalisedIndex weightHarness / source
SWE-bench Verifiedest.8%predicted 82
Terminal-Bench 2.0est.8%predicted 83
DeepSWE68.5%6%mini-swe-agent · $4.65/task
FrontierSWE25.9%5%proximus · $109.71/task
CursorBenchest.4%predicted 42
GSO-Benchest.4%predicted 42
METR Time Horizonsest.10%predicted 79
Vending-Bench 2$5,1656%source
PostTrainBench31.5%4%Claude Code
APEX-Agents47.8%8%source
GDPvalest.7%predicted 54
Remote Labor Indexest.5%predicted 6
DeepResearch Benchest.5%predicted 49
OSWorld 2.0est.6%predicted 16
OSWorldest.4%predicted 84
Cybenchest.5%predicted 82
CL-benchest.5%predicted 24

Other benchmarks

BenchmarkCategoryScoreSource
ARC-AGIReasoning94.5%source
ARC-AGI-2Reasoning60.4%source
Chess PuzzlesReasoning37.0%source
CritPtReasoning23.4%source
DTBenchKnowledge91.2%source
FrontierCodeCoding44.2%mini-swe-agent
FrontierMath Tiers 1-3Math72.2%source
FrontierMath Tier 4Math39.0%source
GPQA DiamondReasoning93.1%source
LMCAKnowledge52.7%source
MirrorCodeCoding54.4%source
OTIS Mock AIMEMath97.2%source
ProofBenchMath87.0%source
SciCodeCoding59.5%source
SimpleBenchReasoning60.7%source
SimpleQA VerifiedKnowledge50.6%source
WebDev ArenaCoding1674source
WeirdML v3Coding7.5%kimi_code

API hosts

Live prices and uptime for every host serving Kimi K3 via OpenRouter.

HostInputOutputCache readAgent $/MUptime 24hQuant
Wafer$0.28$9.60$0.13$0.60100.0%mxfp4
Relace$0.18$11.01$0.18$0.7099.9%fp4
Chutes$1.85$9.50$0.18$0.8796.9%mxfp4
Wafer$0.15$16.00$0.15$0.91100.0%mxfp4
InferenceNet$0.24$15.00$0.21$0.92100.0%fp4
InferenceNet$0.24$15.00$0.21$0.92100.0%fp4
Wafer$0.19$16.00$0.18$0.94100.0%mxfp4
Sail Research$0.34$16.58$0.17$0.98100.0%fp4
KT$0.45$15.00$0.30$1.0299.1%—
Io Net$1.90$14.00$0.19$1.0999.3%fp4
DigitalOcean$2.55$12.95$0.26$1.19100.0%—
InferenceNet$0.45$18.00$0.35$1.20100.0%fp4
InferenceNet$0.45$18.00$0.35$1.20100.0%fp4
DeepInfra$2.85$14.25$0.28$1.3299.8%mxfp4
Alibaba$3.00$15.00$0.30$1.39100.0%—
Amazon Bedrock$3.00$15.00$0.30$1.39100.0%—
BaseTen$3.00$15.00$0.30$1.3999.7%fp8
Fireworks$3.00$15.00$0.30$1.3999.8%—
Modal$3.00$15.00$0.30$1.39100.0%mxfp4
Moonshot AI$3.00$15.00$0.30$1.3999.4%—
Nebius$3.00$15.00$0.30$1.39100.0%fp4
Parasail$3.00$15.00$0.30$1.39100.0%fp4
River$3.00$15.00$0.30$1.3999.9%—
Together$3.00$15.00$0.30$1.3998.4%—
Makora$1.00$18.00$0.50$1.40100.0%fp4
Amazon Bedrock$3.30$16.50$0.33$1.52100.0%—
Fireworks$4.50$22.50$0.45$2.0896.7%—
Fireworks$4.50$22.50$0.45$2.0898.9%—
Parasail$4.50$22.50$0.45$2.08100.0%fp4