Skip to main content

GPT-5.6 Luna on BlueBench

GPT-5.6 Luna ranks 20th of 26 models on BlueBench, Cotool's benchmark for defensive security work, with an overall score of 62.4. It ranks highest in incident response (61.8, 18th of 26) and lowest in threat hunting (52.4, 20th of 26). It averages $0.58 per task, the 6th lowest cost of the 26 models tested.

OpenAI · Closed weight · Updated Oct 2026
Methodology
BlueBench score
62.4
#20 of 26 models
Cost per task
$0.58
#6 cheapest of 26
Reliability
33.4
#16 of 26 models
Best-ranked track
#18
Incident Response · 61.8
Read the GPT-5.6 Luna rundown →Verdict, strengths and weaknesses against similar-cost models, and when to pick something else.

Scores by track

BlueBench weights incident response, threat hunting, and detection engineering at 30% each and malware analysis at 10%.

TrackScoreRankTrack leader
Incident Response61.8#18 of 26GPT-6 Astra · 88.0
Threat Hunting52.4#20 of 26Claude Opus 5.5 · 76.1
Detection Engineering67.5#19 of 26GPT-6 Astra · 87.3
Malware Analysis79.1#12 of 26Claude Opus 5.5 · 89.4

Scores by benchmark

Each BlueBench benchmark is a separate intrusion, real or simulated, with its own tasks and grading.

Models ranked nearby

The models closest to GPT-5.6 Luna on the overall leaderboard.

ModelRankScoreCost / task
#1864.1$0.26
#1962.7$3.01
GPT-5.6 Luna
#2062.4$0.58
#2158.6$0.17
#2256.4$0.90

Frequently asked questions

How good is GPT-5.6 Luna at security operations?
GPT-5.6 Luna scores 62.4 overall on BlueBench and ranks 20th of 26 models on incident response, threat hunting, detection engineering, and malware analysis. The top model, Claude Opus 5.5, scores 83.2.
Which security tasks is GPT-5.6 Luna best at?
BlueBench track scores for GPT-5.6 Luna, best rank first. Malware Analysis: 79.1 (12th of 26); Incident Response: 61.8 (18th of 26); Detection Engineering: 67.5 (19th of 26); Threat Hunting: 52.4 (20th of 26).
How much does GPT-5.6 Luna cost per security task?
GPT-5.6 Luna averages $0.58 per BlueBench task, including repeated attempts. That is the 6th lowest cost of 26 models. The best-scoring cheaper model is DeepSeek V4.1 Flash at $0.17 per task with an overall score of 71.5.
How was GPT-5.6 Luna tested?
Every model runs the same 74 tasks across 6 BlueBench benchmarks, built from real intrusion logs and simulated enterprise environments, using the same harness, prompts, tools, and number of trials. Reports are graded against hidden rubrics and detection rules are re-run against the evidence.

Full leaderboards

AI for the blue team.

Run Cotool's harness in your environment to get real security work done

Book a demo