JetBrains · Model family
This description concerns Mellum-4B. The family also includes Mellum2, described as 12B MoE with 2.5B active parameters for routing, RAG and subagents. The later Mellum2.1 page was inaccessible. https://blog.jetbrains.com/ai/2026/06/mellum2-goes-open-source-a-fast-model-for-ai-workflows/
Choose an exact release and deployment. Account region, data use, retention and service prices depend on that deployment; model evaluations apply only to their stated scope.
The provider’s own sources say different things here. Contradiction over time: 2024 launch post calls Mellum “a proprietary large language model”; open-sourced 2025. HF model card https://huggingface.co/JetBrains/Mellum-4b-base
Provider benchmark charts; evaluate on your own workload.
No Git host/CI integrations (model)
No AI Pro price read (not in scope pages)
Concerns collected telemetry data, not inference hosting region for cloud completion; no inference region found
The provider’s own sources say different things here. Contradiction: data-collection notice says individual non-commercial users are opted in by default (can opt out); data-handling page says setting is off by default. Model trained “solely on publicly available, permissively licensed code” (https://blog.jetbrains.com/blog/2024/10/22/introducing-mellum-jetbrains-new-llm-built-for-developers/)
No subprocessor list reviewed
Applies to JetBrains AI telemetry; self-hosted weights keep data under user control
signals only
Editable hypothetical example · not provider pricing or measured performance. Use this monthly worksheet to test your own quote and quality assumptions.
/ accepted outcome
accepted outcomes / month
5,000 tasks × 80% = 4,000 accepted outcomes per month. Higher entered costs raise unit cost; more accepted outcomes spread the same cost over more results.
Formula: entered monthly costs ÷ (monthly tasks × acceptance % ÷ 100). The numerator includes only the costs you enter. Add applicable licence, usage, integration, human review and rework costs yourself; this is not automatically a fully loaded estimate or a vendor billing model. Capacity released is not realized financial value.
Inputs stay in this browser tab’s session, including reload and language changes. They are not added to the page URL. Continuing prepares an unsent enquiry draft.
Each mark shows what kind of source backs a fact. Open the fact to read the source, the date it was checked and any conditions.
Figures marked 'Scenario' are calculated from stated assumptions. Results measured in practice are shown separately.
Each source supports its recorded scope. Documentation depth and independent verification are separate.
AI DECISION BRIEF
You get the facts that matter for your decision, each with its source, the questions to ask each vendor and a cost estimate based on your own numbers. First you agree your three most important criteria with Sten Franke.
We deliver your brief by email within a maximum of five working days after payment is confirmed and we have agreed the question and minimum information needed to begin. ETHORITY confirms the start date and delivery date and delivers earlier when the reviewed brief is ready. Missing optional information is recorded and assessed; it does not require you to research the solution yourself. Review before delivery and email support are included.
The buttons open an email draft you can edit before sending. You agree the work and timing directly with Sten Franke.
ETHORITY®