Jamba 1.6 Mini

AI21 Labs·Released 2025-03-06

Open Source

Model Family

Jamba Reasoning 3B2025-10-08 AI21: Jamba Large 1.72025-08-08 Jamba 1.7 Large2025-07-07 Jamba 1.7 Mini2025-07-07 Jamba 1.6 Large2025-03-06 Jamba 1.5 Large2024-08-22 Jamba 1.5 Mini2024-08-22

Quality Index

7.9

413th of 442

Top 93%

Price/1M

$0.25

309th cheapest

19% below median

Top 46%

Speed

187 tok/s

Top 9%

TTFT

0.60s

Market Position

Jamba 1.6 MiniMarket Average

Pricing

Input

$0.20

per 1M tokens

Output

$0.40

per 1M tokens

Blended

$0.25

per 1M tokens

Cheaper than 54% of models. Median price is $0.31/1M tokens.

Cost Calculator

Tokens per day1M

100K100M

Daily

$0.25

Monthly

$7.50

vs. Similar Models

Granite 4.0 H 1BQ:+0.1

$0.00-100%

Jamba 1.5 MiniQ:+0.1

$0.250%

LFM2 2.6BQ:+0.1

$0.00-100%

LFM2.5-1.2B-InstructQ:+0.1

$0.00-100%

Performance

187

tokens/sec

Faster than 91% of models

0.60

seconds

Faster than 40% of models

0.60

seconds

Faster than 45% of models

Market Median

46 tok/s

309% faster

Median TTFT

0.42s

44% slower

Throughput/Dollar

746

tok/s per $/1M

Speed Comparison

Gemini 3 Flash Preview (Non-reasoning)

188 tok/s+1%

Devstral Small (Jul '25)

188 tok/s+1%

Gemini 3 Flash Preview (Reasoning)

185 tok/s-1%

Benchmarks

MMLU-Pro

36.7%

GPQA Diamond

30.0%

HLE

4.6%

LiveCodeBench

7.1%

SciCode

10.1%

TerminalBench HardNot evaluated

MATH-500

25.7%

AIME

3.3%

AIME 2025Not evaluated

IFBenchNot evaluated

Long Context RecallNot evaluated

Tau2Not evaluated

Market AverageTop Score

Open Source

Quick Compare

Similar Models

LFM2 2.6B

Liquid AI

Q: 8.0$0.00/1M

Slower: 100%Cheaper: 100%

LFM2.5-1.2B-Instruct

Liquid AI

Q: 8.0$0.00/1M

Slower: 100%Cheaper: 100%

Granite 4.0 H 1B

IBM

Q: 8.0$0.00/1M

Slower: 100%Cheaper: 100%

Jamba 1.5 Mini

AI21 Labs

Q: 8.0$0.25/1M

Slower: 100%

Qwen3 1.7B (Reasoning)

Alibaba

Q: 8.0$0.40/1M

Slower: 25%Pricier: 59%

LFM2.5-1.2B-Thinking

Liquid AI

Q: 8.1$0.00/1M

Slower: 100%Cheaper: 100%

Compare all 7 models