LLM Reference

Mellum2 12B

Released
2026-06-01
Last refreshed
2026-06-29
Status
Researched 45d ago
Open sourceCommercial use: permittedCodingRAGAgentsLong contextJSON / Tool use

Mellum2 12B is a released coding, rag, and agents model with open-source and 131k context; evaluate it while provider pricing coverage matures.

Use it for

  • Teams evaluating coding, rag, and agents
  • Workloads that can use a 131k context window

Do not use it for

  • Cost-sensitive launches that need sourced token pricing
  • Vision or document-understanding workloads
  • Teams that need a tracked hosted API route today
Specifications
Family
Mellum 2
Released
2026-06-01
Context
131k
Parameters
12B (2.5B active)
Architecture
Mixture of Experts
Specialization
code
Openness
Open source
License
Apache 2.0OSI-approvedCommercial use: permitted
Weights
Available
Code
Unknown
Training
Pretrained
Created by

Developer tools company building code-specialized Mellum models.

Prague, Czech Republic
Founded 2000
Website
Pricing

No tracked provider token pricing is available yet.

About

Mellum2 12B is JetBrains' open-weights mixture-of-experts model for software engineering workflows. It has 12B total parameters, activates about 2.5B parameters per token, and supports long-context coding, tool use, function calling, reasoning, and agentic assistance.

Mellum2 12B is an open-source model in the Mellum 2 family. The structured metadata tracks a 131k-token context window, reasoning, function calling, and tool use. Headline tracked benchmarks include Berkeley Function Calling Leaderboard v3 66.3 and HumanEval 78.4.

Top use-case fit: coding, agents, and build tasks

Coding

1 relevant benchmark in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Agents

1 relevant benchmark in the decision map.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Capabilities

ReasoningFunction CallingTool Use

Benchmark peer barsfor Coding

Benchmark scores(2)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionSource
Berkeley Function Calling Leaderboard v366.3BFCL v3 (accuracy%)https://huggingface.co/blog/JetBrains/mellum2-launch
HumanEval78.4HumanEval (pass@1)https://huggingface.co/blog/JetBrains/mellum2-launch

Migration checks

No linked migration route is available for this model yet.

Frequently asked questions

What is the context window of Mellum2 12B?

Mellum2 12B has a context window of 131k tokens.

When was Mellum2 12B released?

Mellum2 12B was released on 2026-06-01.

What benchmarks has Mellum2 12B been tested on?

Mellum2 12B has been evaluated on 2 benchmarks, including Berkeley Function Calling Leaderboard v3, HumanEval.