Everything about Jamba Mini
Jamba Mini in numbers: price per token, context window, independent benchmark scores and comparable models, based on public data checked on October 22, 2026.
Data as of October 22, 2026
Key figures
| Developer | AI21 Labs |
|---|---|
| Released | January 8, 2026 |
| Knowledge cutoff | August 22, 2024 |
| Context window | 256,000 tokens (≈ 384 pages) |
| Maximum output | 4,096 tokens |
| Input price | $0.20 per million tokens (developer's price) |
| Output price | $0.40 per million tokens |
| Open weights | Yes |
| Reasoning mode | No |
| Tool calling | Yes |
| Structured output (JSON) | Yes |
| Accepted inputs | text |
| Providers offering it | 1 |
Overview
Released on January 8, 2026, Jamba Mini is one of the 1 AI21 Labs models tracked on this site. Its weights are published: you can download it and run it on your own servers, or use one of the 1 API providers that list it on models.dev. It has no declared reasoning mode: it answers directly, without an extra billed thinking step. That makes it 9 months old as of October 22, 2026.
We file it under open-weight models (68 models in this category on the site).
What it costs
Jamba Mini costs $0.20 per million input tokens and $0.40 per million output tokens (developer's price, checked on October 22, 2026).
To put it in perspective: reading 20,000 tokens (about thirty pages) and writing 2,000 costs roughly $0.005; a thousand of those tasks, $4.8.
That is 2.5 times cheaper than the median output price of open-weight models ($1).
To estimate your own budget, use the agent cost calculator.
Measured performance
As of October 22, 2026, Epoch AI publishes no score for Jamba Mini on our basket's benchmarks. To place it, rely on its specifications and price, then test it on your own cases.
What it can do
Its context window reaches 256,000 tokens, roughly 384 pages of text at 500 words per page: enough for a long report or several contracts at once. A single answer can run to 4,096 tokens.
It accepts as input: text. Its knowledge stops at August 22, 2024: beyond that, it needs web search or your documents.
It can call tools (functions, APIs) and return structured JSON output, which makes it a possible foundation for an AI agent.
Who is it for?
Given its category, Jamba Mini is mainly aimed at projects where you want control over your data or hosting, or want to adapt the model to your field.
Its very low cost (under half a cent for our sample task) means it can be called thousands of times a day.
Before adopting it, test it on your own examples: benchmarks measure standard tasks, not your job.
Alternatives
In the same category, the models closest to Jamba Mini are:
- Gemma 4 31B IT (Google DeepMind, $0.40 per million output tokens)
- DeepSeek V3.2 (DeepSeek, index 59.8)
- Gemma 4 26B A4B IT (Google DeepMind, $0.34 per million output tokens)
- Nemotron 3.5 Super VL 120B A12B (NVIDIA, $0.50 per million output tokens)
Similar models
- Gemma 4 31B IT · Google DeepMind
- DeepSeek V3.2 · DeepSeek · index 59.8
- Gemma 4 26B A4B IT · Google DeepMind
- Nemotron 3.5 Super VL 120B A12B · NVIDIA
Frequently asked questions
How much does Jamba Mini cost?
As of October 22, 2026, Jamba Mini costs $0.20 per million input tokens and $0.40 per million output tokens (developer's price). A task that reads 20,000 tokens and writes 2,000 costs about $0.005.
What is the context window of Jamba Mini?
256,000 tokens, roughly 384 pages of text. An answer can reach 4,096 tokens.
Is Jamba Mini open source?
Its weights are published: you can install it on your own servers. Check the license before commercial use.
When was Jamba Mini released?
Jamba Mini was released on January 8, 2026, according to models.dev.
Can you build an agent with Jamba Mini?
Yes: it can call tools and return structured JSON, the two basic building blocks of an agent.
Data sources
- Models.dev — Model data: models.dev (open-source project)
- OpenRouter — Market prices cross-checked with OpenRouter
Data as of October 22, 2026
Report an error
Does a figure, a date or a fact look wrong? Tell us: every report is checked.