Granite 4.2 8B vs Ministral 3 3B 2512

Price, context window & benchmark comparison

I

Granite 4.2 8B

IBM Granite
$0.06 / $0.25
Input / Output · per 1M tokens
VS
M

Ministral 3 3B 2512

Mistral AI
$0.10 / $0.10
Input / Output · per 1M tokens
Compare different models
I
M
Verdict: Ministral 3 3B 2512 is 7% cheaper than Granite 4.2 8B on a typical 3:1 input/output mix. Granite 4.2 8B scores higher on the Artificial Analysis Intelligence Index (11.1 vs 4.8).
Granite 4.2 8BMinistral 3 3B 2512
Input price / 1M$0.06$0.10
Output price / 1M$0.25$0.10
Blended price (3:1)$0.11$0.10
Cached input / 1M$0.015$0.01
Context window131K131K
Max output118K105K
Intelligence index11.14.8
Coding index22.44.8
Providers
Best provider uptime (30m)
Vision inputNoYes
ReasoningYesNo
Tool callingYesYes
Released2026-08-312025-12-02
Retirement

Cost calculator

FAQ

Which is cheaper, Granite 4.2 8B or Ministral 3 3B 2512?

Ministral 3 3B 2512 is 7% cheaper than Granite 4.2 8B on a typical 3:1 input/output mix.

Granite 4.2 8B vs Ministral 3 3B 2512: which has a bigger context window?

Granite 4.2 8B: 131,072 tokens. Ministral 3 3B 2512: 131,072 tokens.

How much would 1 million requests cost?

At 2,000 input + 500 output tokens per request: Granite 4.2 8B ≈ $245.00, Ministral 3 3B 2512 ≈ $250.00.

More comparisons