# Infrastructure as code: what AI models recommend, September 2026

IT AI Recommendation Index, September 2026 Edition. Asked as "infrastructure-as-code tool" and as "IaC platform for provisioning cloud resources", six framings each, to twelve models with search on, on behalf of a mid-market B2B company. Page: https://it-ai-index.com/cloud/infrastructure-as-code/

**Standing:** Terraform leads with 46% of first choices; verdict clear leader. 56 first choices across the direct, paraphrase, budget and scale prompts.

## First-choice share

| # | Product | Share | Negative rate | Labels |
|---|---|---|---|---|
| 1 | Terraform | 46% | 20% | 70 |
| 2 | OpenTofu | 39% | 0% | 49 |
| 3 | Pulumi | 7% | 6% | 66 |
| 4 | AWS CDK | 2% | 11% | 28 |
| 5 | Ansible | 0% | 8% | 39 |
| 6 | AWS CloudFormation | 0% | 16% | 38 |
| 7 | Azure Bicep | 0% | 5% | 21 |
| 8 | Crossplane | 0% | 12% | 25 |
| 9 | Spacelift | 0% | 10% | 10 |
| 10 | Azure Resource Manager templates | 0% | 21% | 14 |
| 11 | Chef | 0% | 69% | 13 |
| 12 | Puppet | 0% | 69% | 16 |

## Each model's first choice on the direct prompt

- Claude Haiku 4.5: Terraform; alternatives AWS CloudFormation, Ansible, Pulumi
- GPT-5.4 mini: OpenTofu, Terraform; alternatives AWS CloudFormation, Azure Bicep, Pulumi
- Gemini 3.5 Flash: OpenTofu; alternatives Pulumi, Spacelift, Terraform, env0
- Perplexity Sonar: Terraform; alternatives Ansible, OpenTofu, Pulumi
- Grok 4.1 Fast: OpenTofu, Terraform; alternatives AWS CloudFormation / CDK, Atlantis, Pulumi
- Mistral Small: Terraform; alternatives AWS CDK, Azure Bicep, OpenTofu, Pulumi
- DeepSeek V4 Flash: Terraform; alternatives AWS CloudFormation, Ansible, OpenTofu, Pulumi
- Llama 4 Maverick: no first choice
- Qwen 3.7 Flash: Terraform; alternatives AWS CDK, Ansible, Azure Bicep, Pulumi
- Kimi K2: OpenTofu, Terraform; alternatives AWS CDK, Pulumi
- GLM 4.7 FlashX: OpenTofu, Terraform; alternatives AWS CDK, AWS CloudFormation, Pulumi
- MiniMax M2.5: Terraform; alternatives AWS CloudFormation, Pulumi

## Sources the answers cite

57 of 72 answers came back with a source list, from 12 of 12 models. Sites named in the most answers:

- pulumi.com: 36 answers, 47 citations
- apiiro.com: 31 answers, 31 citations
- guideflow.com: 26 answers, 26 citations
- cortex.io: 22 answers, 22 citations
- checkthat.ai: 21 answers, 22 citations
- spacelift.io: 20 answers, 23 citations
- devtoollab.com: 16 answers, 16 citations
- firefly.ai: 13 answers, 14 citations

Pages named in the most answers:

- https://apiiro.com/blog/best-iac-tools (31 answers)
- https://pulumi.com/blog/infrastructure-as-code-tools (31 answers)
- https://guideflow.com/blog/infrastructure-as-code-tools (26 answers)
- https://cortex.io/post/best-infrastructure-as-code-iac-tools (22 answers)
- https://checkthat.ai/answers/what-are-the-best-infrastructure-as-code-tools (21 answers)
- https://devtoollab.com/blog/best-infrastructure-as-code-tools (16 answers)
- https://spacelift.io/blog/infrastructure-as-code-tools (14 answers)
- https://computingforgeeks.com/best-infrastructure-as-code-iac-cloud-automation-tools (12 answers)
- https://gitnux.org/best/infrastructure-as-code-software (12 answers)
- https://devopsplatforms.com/buying-guides/best-iac-tools (10 answers)

## Warned against

- Puppet: 11 of 16 labels negative. "Tools Often Considered Legacy or Declining (Be Cautious for New Projects)... Avoid for new work: Pure legacy config tools" (Grok 4.1 Fast, negative prompt)
- Chef: 9 of 13 labels negative. "Chef Infra Server hits end-of-life in Nov 2026... Avoid for new work: Pure legacy config tools" (Grok 4.1 Fast, negative prompt)
- Terraform: 14 of 70 labels negative. "Avoid starting new projects with: 1. Terraform under BSL (use OpenTofu instead)" (Kimi K2, negative prompt)
- SaltStack: 5 of 7 labels negative. "Mutable infrastructure model (in-place changes lead to drift), steep learning curves... declining adoption" (Grok 4.1 Fast, negative prompt)

## Record

- Method: https://it-ai-index.com/methodology/
- Raw judge labels and full responses: https://it-ai-index.com/data/
- License: CC BY 4.0. Cite as IT AI Recommendation Index, September 2026 Edition, it-ai-index.com.