# Coolset vs Persefoni: which do AI models recommend for ESG and carbon reporting, October 2026

Finance AI Recommendation Index, October 2026 Edition, ESG and carbon reporting. One of fourteen models named Coolset first on the direct prompt; zero named Persefoni. Page: https://finance-ai-index.com/corporate/esg-and-carbon-reporting/coolset-vs-persefoni/

| | First-choice share | Rank | Negative rate | Labels | Models naming it |
|---|---|---|---|---|---|
| Coolset | 7% | #3 of 13 | 9% | 11 | 6 of 14 |
| Persefoni | 5% | #4 of 13 | 3% | 30 | 14 of 14 |

## The direct prompt, model by model

- Gemini 3.5 Flash: coolset first (first choices: Coolset, Greenly) (alternatives: Gravity, KEY ESG, Novata, Sweep)
- Claude Haiku 4.5: neither first, one named (first choices: EcoOnline ESG) (alternatives: Coolset, Novata, Sweep)
- DeepSeek V4 Flash: neither first, one named (first choices: Sweep) (alternatives: Coolset, Greenly, Novisto)
- Qwen 3.7 Flash: neither first, one named (first choices: Watershed) (alternatives: Carbon Cloud, Enablon, Persefoni, PlanA, Sphera)
- GLM 4.7 FlashX: neither first, one named (first choices: Workiva) (alternatives: Greenly, Measurabl, Novisto, Persefoni, Position Green)
- Muse Glimmer 30B: neither first, one named (first choices: Novisto, Sweep) (alternatives: Coolset, Novata)
- GPT-5.4 mini: neither named (first choices: Workiva) (alternatives: Credibl ESG, IBM Envizi, Microsoft Cloud for Sustainability)
- Perplexity Sonar: neither named (first choices: Greenly) (alternatives: EcoOnline ESG, Novata, Novisto, Sweep)
- Grok 4.1 Fast: neither named (first choices: Greenly, Sweep) (alternatives: Horizon ESG, cubemos)
- Mistral Small: neither named (first choices: Sweep) (alternatives: EcoOnline ESG, Novata)
- Llama 4 Maverick: neither named (first choices: Sweep) (alternatives: EcoOnline ESG, Novata)
- Kimi K2: neither named (first choices: Novata, Position Green) (alternatives: Greenly, Novisto, Workiva)
- MiniMax M2.5: neither named (first choices: Plan A, Sweep) (alternatives: EcoOnline ESG, Novata)
- GPT-6 Luna: neither named (first choices: Greenly) (alternatives: Novisto, Workiva)

## What the models said about Coolset

- "Generic/high emission factors until customized." (Grok 4.1 Fast, negative prompt, soft negative)
- "I'd pick Coolset for most mid-sized B2B companies because the available guidance specifically points to it as well-suited to the mid-market use case." (Perplexity Sonar, paraphrase prompt, first choice)
- "Mid-market platforms like Coolset, EcoOnline ESG, Sweep, or Guideflow often fit best... Coolset for multi-framework AI" (Grok 4.1 Fast, scale prompt, first choice)
- "Coolset (strong CSRD, EU Taxonomy, EcoVadis coverage)" (DeepSeek V4 Flash, scale prompt, first choice)

## What the models said about Persefoni

- "good for complex organizations, but may be overkill for mid-sized" (GLM 4.7 FlashX, paraphrase prompt, soft negative)
- "Choose Watershed or Persefoni if: Your primary immediate pain point is calculating Scope 1, 2, and 3 emissions accurately" (Qwen 3.7 Flash, comparative prompt, first choice)
- "Start for free with carbon + basic reporting: Persefoni Pro." (Muse Glimmer 30B, budget prompt, first choice)
- "Best overall for a mid-sized B2B company: Persefoni" (GPT-5.4 mini, paraphrase prompt, first choice)

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. Comparisons are drawn for the top eight products in each category. Published under CC BY 4.0; the output is the models' output, and nothing here is a recommendation by the index.