Granite 4.2 8B vs Nemotron 3 Super

Price, context window & benchmark comparison

I

Granite 4.2 8B

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

Nemotron 3 Super

NVIDIA
$0.08 / $0.45
Input / Output · per 1M tokens
Compare different models
I
N
Verdict: Granite 4.2 8B is 38% cheaper than Nemotron 3 Super on a typical 3:1 input/output mix. Nemotron 3 Super scores higher on the Artificial Analysis Intelligence Index (12.8 vs 11.1). Nemotron 3 Super has the larger context window (262K vs 131K).
Granite 4.2 8BNemotron 3 Super
Input price / 1M$0.06$0.08
Output price / 1M$0.25$0.45
Blended price (3:1)$0.11$0.17
Cached input / 1M$0.015
Context window131K262K
Max output118K236K
Intelligence index11.112.8
Coding index22.437.7
Providers
Best provider uptime (30m)
Vision inputNoNo
ReasoningYesYes
Tool callingYesYes
Released2026-08-312026-03-11
Retirement

Cost calculator

FAQ

Which is cheaper, Granite 4.2 8B or Nemotron 3 Super?

Granite 4.2 8B is 38% cheaper than Nemotron 3 Super on a typical 3:1 input/output mix.

Granite 4.2 8B vs Nemotron 3 Super: which has a bigger context window?

Granite 4.2 8B: 131,072 tokens. Nemotron 3 Super: 262,144 tokens.

How much would 1 million requests cost?

At 2,000 input + 500 output tokens per request: Granite 4.2 8B ≈ $245.00, Nemotron 3 Super ≈ $385.00.

More comparisons