GrowthBook vs Split
Three of fourteen models named GrowthBook first on the direct prompt; zero named Split. GrowthBook was named by fourteen of the fourteen models and Split by twelve and GrowthBook carries 47 labels and Split 26, so the shares are not directly comparable.
Named in one category this edition.
Named in one category this edition.
Share is the count of first choices across the direct, paraphrase, budget and scale prompts over all fourteen models, for a mid-market B2B company; rank is within the category; every quote names the model and the prompt it came from. Both figures come from the feature flags page.
By framing
GrowthBookFirst choices, of fourteen modelsSplit
Across every category in the November 2026 Edition, GrowthBook and Split were named in the same answer forty-nine times, of the 127 answers naming GrowthBook and the 80 naming Split. In those answers Split took the first choice one time and GrowthBook five.
Every model, every framing
| Model | Direct | Paraphrase | Comparative | Budget-constrained | Scale-constrained | Negative |
|---|---|---|---|---|---|---|
| Claude Haiku 4.5 | ||||||
| GPT-5.4 mini | | | | |||
| Gemini 3.5 Flash | | | | |||
| Perplexity Sonar | | | | | ||
| Grok 4.20 | | | | | ||
| Mistral Small | | | ||||
| DeepSeek V4 Flash | | | | | ||
| Llama 4 Maverick | ||||||
| Qwen 3.7 Flash | | | | | ||
| Kimi K2 | | | | | ||
| GLM 4.7 FlashX | | | | | ||
| MiniMax M2.5 | | | | | ||
| GPT-6 Luna | | | | |||
| Muse Glimmer 30B | | | | |
GrowthBook Split first choice named as an alternative argued againstblank: not namedEach cell is one answer, GrowthBook on the left and Split on the right.The direct prompt
GrowthBook first, Split an alternative
3 of 14 modelsSplit was named in the answer but not as the choice, or not at all.Neither was the first choice, one was named
6 of 14 modelsThe answer put something else first and named one of the two as an alternative.Neither was named
5 of 14 modelsThe answer made no first choice from these two in this category.Bold names in an answer are the products the judge labeled a first choice; a model naming several gives each of them that label. The full answer text for every row is in the record.
By buyer segment
What the models said about GrowthBook
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Five of six in this category shown.
“Any self-hosted option—including Unleash, Flagsmith, or GrowthBook: be cautious” GPT-6 Luna · negative prompt · soft negative
“GrowthBook — full SaaS without viable self-hosting option” DeepSeek V4 Flash · negative prompt · soft negative
“For most mid-market B2B companies, GrowthBook or Flagsmith provide the best balance of features, pricing, and ease of use.” GLM 4.7 FlashX · direct prompt · first choice
“GrowthBook stands out as the best overall value and capability match for mid-market B2B companies.” DeepSeek V4 Flash · direct prompt · first choice
“My default pick: GrowthBook for a small company that wants to keep costs near zero.” GPT-6 Luna · budget prompt · first choice
What the models said about Split
Every negative label with a quote, up to three, then the highest-weighted positives, up to three. Six of eight in this category shown.
“Avoid paid-heavy tools like LaunchDarkly, Split.io, or DevCycle” Grok 4.20 · budget prompt · hard negative
“Be cautious about opaque, quote-based pricing, significant switching costs, and coupling of feature flagging to a broader Harness toolchain if you want portability.” Muse Glimmer 30B · negative prompt · soft negative
“more expensive than alternatives for comparable functionality, lacks automatic analytics built into every flag check” Qwen 3.7 Flash · negative prompt · soft negative
“High-End/Enterprise: LaunchDarkly, Split.io, Optimizely.” MiniMax M2.5 · scale prompt · first choice
“Choose Split/Harness if experimentation is the core use case and you're comfortable with a more enterprise-oriented tool” GPT-5.4 mini · paraphrase prompt · alternative
“Choose Statsig, GrowthBook, or Split/Harness when measuring product impact and running experiments is central.” Perplexity Sonar · comparative prompt · alternative
Comparisons are drawn for the top eight products in each category, each against each. The output is the models' output; nothing here is a recommendation by the index.