SShogo Bench

GPT-6.1 Terra vs Kimi K3

GPT-6.1 Terra leads Kimi K3 by 14.1 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. GPT-6.1 Terra wins 11 and Kimi K3 wins 1 of 12 shared benchmarks. GPT-6.1 Terra is cheaper for agent workloads.

Add more models to this comparison →

Pillars

PillarGPT-6.1 TerraKimi K3
Agentic coding7963
Long-horizon autonomy7465
Knowledge-work agents5642
Computer use6343
Security & context learning6253

Shared benchmarks

BenchmarkGPT-6.1 TerraKimi K3
APEX-Agentsindex68.3%47.8%
Vending-Bench 2index$9,107$5,165
CritPt28.0%23.4%
FrontierMath Tier 492.7%39.0%
FrontierMath Tiers 1-390.9%72.2%
GPQA Diamond94.3%93.1%
MirrorCode76.7%54.4%
OTIS Mock AIME100.0%97.2%
SciCode55.1%59.5%
SimpleBench72.2%60.7%
SimpleQA Verified57.3%50.6%
WeirdML v328.1%7.5%

Specs and price

GPT-6.1 TerraKimi K3
Agent Index69 (#8)54.9 (#26)
Agent $/M$0.42$1.39
Input / output $/M$1.00 / $5.00$3.00 / $15.00
Context1.1M1M
ReleasedOct 26, 2026Jul 16, 2026
WeightsClosedOpen
API hosts429

Don’t pick one. Route.

Shogo can run GPT-6.1 Terra and Kimi K3 side by side and send each step of an agent to the one that’s cheaper for it.

Try both in Shogo