Kimi K3 on BlueBench
Kimi K3 ranks 16th of 26 models on BlueBench, Cotool's benchmark for defensive security work, with an overall score of 67.3. It ranks highest in threat hunting (64.3, 8th of 26) and lowest in detection engineering (62.0, 24th of 26). It averages $2.62 per task, the 17th lowest cost of the 26 models tested. It is the 4th highest-scoring of 8 open-weight models.
Moonshot AI · Open weight · Updated Oct 2026
Methodology - BlueBench score
- 67.3
- #16 of 26 models
- Cost per task
- $2.62
- #17 cheapest of 26
- Reliability
- 36.3
- #15 of 26 models
- Best-ranked track
- #8
- Threat Hunting · 64.3
Scores by track
BlueBench weights incident response, threat hunting, and detection engineering at 30% each and malware analysis at 10%.
| Track | Score | Rank | Track leader |
|---|---|---|---|
| Incident Response | 72.3 | #12 of 26 | GPT-6 Astra · 88.0 |
| Threat Hunting | 64.3 | #8 of 26 | Claude Opus 5.5 · 76.1 |
| Detection Engineering | 62.0 | #24 of 26 | GPT-6 Astra · 87.3 |
| Malware Analysis | 77.2 | #13 of 26 | Claude Opus 5.5 · 89.4 |
Scores by benchmark
Each BlueBench benchmark is a separate intrusion, real or simulated, with its own tasks and grading.
| Benchmark | Score | Rank |
|---|---|---|
| Simulation-003Impact & Exfiltration | 62.8 | #18 of 26 |
| Simulation-002Identity & Active Directory Attacks | 60.2 | #16 of 26 |
| Simulation-001Initial Access & Command-and-Control | 57.5 | #8 of 26 |
| Intrusion-003AWS Cloud Intrusion | 79.9 | #10 of 26 |
| Intrusion-002Windows Enterprise Intrusion | 76.0 | #7 of 26 |
| Intrusion-001macOS Threat Investigation | 75.1 | #22 of 26 |
Models ranked nearby
The models closest to Kimi K3 on the overall leaderboard.
| Model | Rank | Score | Cost / task |
|---|---|---|---|
| #14 | 69.9 | $1.01 | |
| #15 | 68.5 | $0.59 | |
Kimi K3 | #16 | 67.3 | $2.62 |
| #17 | 65.5 | $1.03 | |
| #18 | 64.1 | $0.26 |
Frequently asked questions
- How good is Kimi K3 at security operations?
- Kimi K3 scores 67.3 overall on BlueBench and ranks 16th 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 Kimi K3 best at?
- BlueBench track scores for Kimi K3, best rank first. Threat Hunting: 64.3 (8th of 26); Incident Response: 72.3 (12th of 26); Malware Analysis: 77.2 (13th of 26); Detection Engineering: 62.0 (24th of 26).
- How much does Kimi K3 cost per security task?
- Kimi K3 averages $2.62 per BlueBench task, including repeated attempts. That is the 17th lowest cost of 26 models. The best-scoring cheaper model is Claude Sonnet 5.5 at $1.01 per task with an overall score of 77.5.
- Is Kimi K3 the best open-weight model for security work?
- No. Kimi K3 ranks 4th of 8 open-weight models. DeepSeek V4.1 Flash leads the open-weight models with 71.5.
- How was Kimi K3 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