LLM·Dex
Open weightsAI21smalltext

Jamba 1.5 Mini

Smaller hybrid SSM-Transformer model, fast and efficient at long contexts.

Updated


Quick facts

Released
Aug 2024
Context
256K tokens
Output / 1M
$0.40 / 1M tokens
License
Jamba Open Model License

About Jamba 1.5 Mini

Jamba 1.5 Mini brings AI21's hybrid Mamba-Transformer architecture to a smaller parameter count, useful for cost-sensitive long-context workloads. The 256k context window is the standout feature at this price tier, and the Mamba state-space components keep memory and throughput economics favourable on long inputs.

Jamba Mini is most often deployed as the input-stage model in two-tier pipelines that summarise large corpora before handing off to a more capable downstream model.

Benchmarks

Published scores from AI21's model card or independent leaderboards. We do not publish numbers we cannot source, see methodology.

HumanEval
,
Python coding pass@1
MMLU
,
Broad academic knowledge
GPQA
,
Graduate-level reasoning
SWE-bench
,
Real software-engineering tasks
Benchmark scores not yet available. We only publish numbers we can source from official model cards or independent leaderboards, see methodology.

Capabilities

Strengths

  • 256k context at small-model price
  • Hybrid SSM-Transformer is efficient on long inputs
  • Open weights for self-hosting

Tracked weaknesses

  • Quality below mid-tier frontier

Pricing

Per-million-token rates as published by AI21.

TierPriceNotes
Input$0.20 / 1M tokensTokens you send to the model
Output$0.40 / 1M tokensTokens the model generates
Context256K tokensMax combined input + output

Call Jamba 1.5 Mini from your code

Drop-in snippet for the AI21 SDK. Set your API key in the environment and run.

typescript
import OpenAI from "openai";

const client = new OpenAI({
  // Use OPENAI_API_KEY for OpenAI, or your provider's key + baseURL.
  apiKey: process.env.OPENAI_API_KEY,
});

const completion = await client.chat.completions.create({
  model: "jamba-1-5-mini",
  messages: [
    { role: "user", content: "What's the time complexity of quicksort?" },
  ],
});

console.log(completion.choices[0].message.content);

Compare Jamba 1.5 Mini with…

Frequently asked

  • How much does Jamba 1.5 Mini cost per million tokens?
    Jamba 1.5 Mini is priced at $0.20 / 1M tokens for input tokens and $0.40 / 1M tokens for output tokens via the official AI21 API at the time of writing.
  • What is Jamba 1.5 Mini's context window?
    Jamba 1.5 Mini supports a context window of 256K tokens.
  • Is Jamba 1.5 Mini open source?
    Jamba 1.5 Mini ships with open weights under the Jamba Open Model License license. You can self-host it, fine-tune it, and (subject to the license terms) deploy it commercially.
  • When was Jamba 1.5 Mini released?
    Jamba 1.5 Mini was released on Aug 22, 2024 by AI21.
Friday digest

Intelligence, distilled weekly.

One short email every Friday, new model launches, leaderboard moves, and pricing drops. Curated by hand. Free, no spam.