Kimi K3.1 Pro on iac-bench 2026-10 Moonshot AI's Kimi K3.1 Pro scored an overall composite of 86.5 (95% CI 84.8–88.7) on the iac-bench 2026-10 infrastructure-as-code benchmark, ranking sixth of 16 models and the highest open-weight model in the field, according to the benchmark's published results. Kimi K3.1 Pro ranked third on the mechanical score alone (85.6, behind Claude Fable 5.2 and Claude Opus 5.5) and ninth on the judge score (88.6), with all 9 attempts planning successfully, zero failed gates and zero security findings, at an estimated total cost of $3.52. The model's strongest fixture was S3 (G001), where it tied fourth by median composite at 85.6, while App Service (G003) had its highest median (87.8) but placed eighth and Cloud SQL (G002) placed seventh (86.0, up from 79.2 for Kimi K3). New: LLM Rankings are live See how the top LLMs rank on real infrastructure tasks https://www.massdriver.cloud/llm-rankings How Kimi K3.1 Pro did on iac-bench 2026-10: an overall composite of 86.5, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules. Every model evaluated on the 2026-10 fixtures side by side, with a sortable leaderboard, picks by use case and the cost of each: LLM rankings https://www.massdriver.cloud/llm-rankings . | Developer | Moonshot AI | | API identifier | moonshotai/kimi-k3.1-pro | | Access | OpenRouter, pinned to Moonshot AI's endpoint MXFP4 | | Weights | Open | | List price USD / M tokens, in / out | 2.00 / 10.00 OpenRouter, 2026-11-10 | | Evaluated | 2026-11-10 generation and judge | | Attempts | 9 3 fixtures × 3 samples | | Overall composite | 86.5 95% CI 84.8–88.7 | | Rank | 6 of 16 95% rank interval 3–9 | | Mechanical only / judge only | 85.6 rank 3 / 88.6 rank 9 | | Failed gates · insecure attempts | 0/9 · 0/9 | | Estimated cost, all 9 attempts | $3.52 | Kimi K3.1 Pro places sixth of 16 by composite 86.5 , the highest open-weight model, with a 95% rank interval of 3–9. It is third on the mechanical score alone 85.6, behind Claude Fable 5.2 and Claude Opus 5.5 and ninth on the judge score 88.6 . In the bootstrap it finishes ahead of Claude Sonnet 5.5 in 54% of resamples, ahead of Claude Fable 5.1 in 35%, GPT-6.1 Sol in 33% and GPT-6 Astra in 27%, and ahead of Kimi K3 in more than 99%. All 9 attempts cost $3.52, about the same as Claude Opus 5.5 $3.58 , against $1.52 for Sonnet, $1.02 for Sol and $3.96 for Kimi K3. All 9 attempts planned, with no security findings. Kimi K3 had four findings and one failed gate. S3 G001 is its strongest fixture, tied fourth in the field by median composite 85.6, with Claude Fable 5.1 . App Service G003 has its highest median 87.8 but places eighth, and Cloud SQL G002 places seventh 86.0, up from 79.2 for Kimi K3 . | Fixture | Composite median range | Rubric median | Expertise | Judge median | ✗ gate | ⚠ security | Median inputs | Median output tokens | Median time | Median cost | |---|---|---|---|---|---|---|---|---|---|---| | G001 · S3 | 85.6 84.0–86.4 | 0.849 | 4/7 | 92.1 | 0/3 | 0/3 | 12 | 12,640 | 526 s | $0.174 | | G002 · Cloud SQL | 86.0 84.2–91.8 | 0.875 | 7/9 | 86.8 | 0/3 | 0/3 | 24 | 19,216 | 720 s | $0.300 | | G003 · App Service | 87.8 84.7–87.8 | 0.889 | 4/5 | 86.8 | 0/3 | 0/3 | 51 | 42,983 | 1866 s | $0.669 | Rubric items not at full credit, counted over attempts that planned: G001: A5 3/3, H12 3/3, H13 3/3, H15 3/3; G002: A8 1/3, B1 3/3, H4 3/3, H5 3/3, H9 1/3, H11 2/3; G003: A7 3/3. Definitions are in Scoring criteria https://www.massdriver.cloud/blogs/iac-bench-2026-10-scoring-criteria . ssl mode = "ENCRYPTED ONLY" is a literal in 3/3 attempts, next to ipv4 enabled = false H2 passes 3/3; Kimi K3 failed it 3/3 by validating an input that admitted unencrypted connections . No S3 attempt offers public reads H1 , which Kimi K3 lost in 1/3 . cloudsql.iam authentication = "on" last, so it cannot be overridden. log connections and log disconnections are on in 3/3 E8, which Kimi K3 misses in 3/3 . health check eviction time in min , the pairing that failed Kimi K3's gate. Autoscale, WEBSITES PORT , metric alerts and AcrPull E1–E4 appear in 3/3; Key Vault references E5 in none. B1 is lost in 3 of 9 attempts, all on Cloud SQL, against 6 of 9 for Kimi K3 and Claude Fable 5.1 and 5 of 9 for GPT-6.1 Sol. A7 scores zero in 3/3. Kimi K3 used 20–34, and the next-largest median is Gemini 3.6 Flash at 40. Many are attribute pass-throughs ip restrictions , vnet route all enabled , container start time limit . container registry id beside container registry url in 3/3, and a managed identity's client ID beside its principal ID in e2. A registry password is accepted as an alternative to .gserviceaccount.com suffix, which Cloud SQL rejects, in 3/3 Kimi K3 trims it in 3/3 . The e1 and e3 READMEs document the full email as the username: H4 / H5 at 0.75, 3/3 , where Kimi K3 offered one; ENTERPRISE PLUS 35 days is offered; Cloud SQL provider ranges such as = 5.20.0, < 7.0.0 that span two majors; no AWS provider block H15 , 3/3 ; S3 over budget with 12–13 inputs A5 , 3/3 . H11 lost 2/3 . deletion policy . e2 defaults it to ABANDON through an optional attribute of the databases map, which earns full credit. H13 , 3/3 . logging.amazonaws.com instead of logging.s3.amazonaws.com excerpt below . cloudsql.instanceUser and cloudsql.client to each IAM user E6 , which Kimi K3 never does. Validation stays as dense as Kimi K3's: 20–21 blocks per Cloud SQL module, 22–28 per App Service module. G001/kimi-k3.1-pro-e2/main.tf lines 254–270 statement { sid = "AllowServerAccessLogDelivery" effect = "Allow" principals { type = "Service" identifiers = "logging.amazonaws.com" } actions = "s3:PutObject" resources = "${aws s3 bucket.logs 0 .arn}/access-logs/ " condition { test = "ArnLike" variable = "aws:SourceArn" values = aws s3 bucket.this.arn } The one S3 attempt that creates its own log bucket also writes a delivery grant for it, scoped to this bucket's ARN, but names the service principal logging.amazonaws.com. S3 server access logs are delivered by logging.s3.amazonaws.com, so the statement admits nothing and the rubric's delivery check H13 fails. How Claude Fable 5.1 did on iac-bench 2026-10: an overall composite of 86.9, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules. How Claude Fable 5.2 did on iac-bench 2026-10: an overall composite of 93.9, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules. How Claude Haiku 5.5 did on iac-bench 2026-10: an overall composite of 68.0, scores per fixture, strengths, weaknesses and the patterns in its OpenTofu modules.