Fourteen AI models were asked for CI/CD platform six ways each, on behalf of a small, a mid-market and an enterprise B2B company: 252 answers, in which a judge labeled 43 products. Four of them carry at least 10 labels and a first choice, and are ranked. This guide walks the top four in the order the mid-market standing puts them, with what the models said for and against each, in their words. GitHub Actions leads at 68% of first choices, a clear leader.
| Product | First-choice share | Negative rate | Labels | Leads for | Quadrant | |
|---|---|---|---|---|---|---|
| 1 | 68% | 7% | 68 | Small business, Mid-market | endorsed leader | |
| 2 | 23% | 7% | 73 | Enterprise | accepted challenger | |
| 3 | 4% | 19% | 64 | none | accepted challenger | |
| 4 | 2% | 21% | 19 | none | accepted challenger |
The full standing, every product with 10 labels or more, is on the category page, with the products at zero and the ones warned against.
A product is on the record because a model named it in an answer and the judge labeled the naming: first choice, alternative, mention, caution or warning, each with the sentence it came from. 43 products were named for CI/CD platform this edition.
A product is ranked once it carries 10 labels in a segment and at least one first choice there. Below that it is named but not ranked, and it says so on its page. The vendor table that folds spellings into products is versioned; this page was scored under v2026-10.8.
The same categories are asked again with nothing changed to measure how much an answer moves on its own. A change in share smaller than the floor is not reported as a change. The repeat, model by model.
| Framing | The question |
|---|---|
| Direct | What is the best CI/CD platform for a mid-market B2B company? |
| Paraphrase | Which continuous integration and delivery tool would you recommend to a mid-sized B2B company? |
| Comparative | What are the top CI/CD platforms and how do they differ? |
| Budget-constrained | What is the best CI/CD platform for a company with a limited budget? |
| Scale-constrained | We are a 500 person company evaluating a CI/CD platform. What should we look at? |
| Negative | Which CI/CD platforms should I avoid or be cautious about? |
GitHub Actions is #1 of 4 for the mid-market buyer at 68% of first choices, from 68 labels by 14 of 14 models; 7% of those labels were cautions or warnings. Ten of fourteen models named it first on the direct question. It led the direct, paraphrase, comparative, budget-constrained, scale-constrained framings. By buyer: small business #1 of 3 at 75%; mid-market #1 of 4 at 68%; enterprise #2 of 8 at 19%.
5 of 14 models argued against it somewhere in their answers, 0 as a warning.
“"GitHub Actions is the right default for the vast majority of teams in 2026, and it's where I'd start unless you have a specific reason not to."” Muse Glimmer 30B · budget prompt · first choice
“If teams are already on GitHub, GitHub Actions is the de facto default: bundled with GitHub Enterprise, 20,000+ marketplace actions.” Muse Glimmer 30B · scale prompt · first choice
“GitHub Actions workflows that trust third-party actions or pull-request code too much” GPT-6 Luna · negative prompt · soft negative
“This is currently the easiest place to start, but it can become the most expensive.” Qwen 3.7 Flash · negative prompt · soft negative
GitLab CI/CD is #2 of 4 for the mid-market buyer at 23% of first choices, from 73 labels by 14 of 14 models; 7% of those labels were cautions or warnings. Two of fourteen models named it first on the direct question. By buyer: small business #2 of 3 at 15%; mid-market #2 of 4 at 23%; enterprise #1 of 8 at 42%.
5 of 14 models argued against it somewhere in their answers, 0 as a warning.
“Some of the top recommendations include: 1. GitLab CI: A comprehensive CI/CD solution integrated within the GitLab DevSecOps platform.” Llama 4 Maverick · paraphrase prompt · first choice
“GitHub Actions or GitLab CI/CD offer the best combination of features, ease of use, and generous free tiers.” Kimi K2 · budget prompt · first choice
“Over 810 hours, caused by faulty updates and overload of CI/CD services during periods of heavy traffic.” Muse Glimmer 30B · negative prompt · soft negative
“GitLab specifically warns about shared non-ephemeral runners and privileged containers” GPT-6 Luna · negative prompt · soft negative
CircleCI is #3 of 4 for the mid-market buyer at 4% of first choices, from 64 labels by 14 of 14 models; 19% of those labels were cautions or warnings. One of fourteen models named it first on the direct question. By buyer: small business #3 of 3 at 5%; mid-market #3 of 4 at 4%; enterprise unranked.
8 of 14 models argued against it somewhere in their answers, 0 as a warning.
“GitHub Actions or CircleCI are the best starting points due to their generous free tiers and ease of use” Mistral Small · budget prompt · first choice
“CircleCI or GitLab CI are the best CI/CD platforms due to their balance of features, scalability, and cost” Mistral Small · direct prompt · first choice
“most companies have migrated to GitHub Actions to avoid paying for two separate platform subscriptions” Gemini 3.5 Flash · direct prompt · soft negative
“solid standalone CI tool, but it's CI-focused (not full CD) and has faced pricing pressure” DeepSeek V4 Flash · paraphrase prompt · soft negative
Buildkite is #4 of 4 for the mid-market buyer at 2% of first choices, from 19 labels by 9 of 14 models; 21% of those labels were cautions or warnings. Zero of fourteen models named it first on the direct question. By buyer: small business unranked; mid-market #4 of 4 at 2%; enterprise #4 of 8 at 11%.
“Buildkite or GitLab CI/CD are excellent choices due to their balance of scalability, ease of use, and integration capabilities” Mistral Small · paraphrase prompt · first choice
“Buildkite or Harness are worth strong consideration if you need high-scale performance or advanced CD capabilities.” DeepSeek V4 Flash · scale prompt · alternative
“Consider Buildkite; it supports hosted and self-managed agents, but self-managed infrastructure adds operational work.” GPT-6 Luna · paraphrase prompt · soft negative
“likely overkill for most mid-sized B2B companies” DeepSeek V4 Flash · paraphrase prompt · soft negative
| Framing | Named first most often | Then |
|---|---|---|
| Direct | GitLab CI/CD (2), CircleCI (1) | |
| Paraphrase | GitLab CI/CD (7), Buildkite (1) | |
| Comparative | GitLab CI/CD (2) | |
| Budget-constrained | CircleCI (1), GitLab CI/CD (1), Podman (1) | |
| Scale-constrained | GitLab CI/CD (3) | |
| Negative | no first choice |
| Buyer | Leads | Then |
|---|---|---|
| Small business | GitLab CI/CD, CircleCI | |
| Mid-market | GitLab CI/CD, CircleCI, Buildkite | |
| Enterprise | GitHub Actions, Azure DevOps, Buildkite |
Across the categories asked twice, a leader's share moved 4 points at the median and the index calls a change only above 11 points. Every category by buyer.
GitHub Actions, in 68% of first choices for a mid-market B2B company in the October 2026 Edition, from 68 labels. The index calls that a clear leader.
Small business: GitHub Actions at 75%. Mid-market: GitHub Actions at 68%. Enterprise: GitLab CI/CD at 42%. Each standing is computed within its segment and never pooled.
The direct, paraphrase, budget and scale framings count toward share. The leader is the same on all six. The table above has the counts.
Asked again with nothing changed, the models moved their own first choice 63% of the time across the repeat sample; Across the categories asked twice, a leader's share moved 4 points at the median and the index calls a change only above 11 points.
Fourteen of the fourteen models return sources. Their 70 answers here cite 875 pages; the sites cited most are blog.jetbrains.com, atlassian.com, dev.to. The category page lists them all.
No. A product is on the record because a model named it. A vendor can claim its page, propose corrections to the vendor table and be told when its standing moves; it cannot change a label, a share or a rank, and the publisher's conflicts are disclosed on the method page.
Every answer, every label and its evidence quote are in the record, sent on request. The page is published under CC BY 4.0. The output is the models' output; nothing here is a recommendation by the index.