Nemotron 3.5 Super VL vs Sonar

Price, context window & benchmark comparison

NVIDIA

Nemotron 3.5 Super VL

NVIDIA
$0.15 / $0.47
Input / Output · per 1M tokens
VS
Perplexity

Sonar

Perplexity
$1.00 / $1.00
Input / Output · per 1M tokens
Compare different models
NVIDIA
Perplexity
Verdict: Nemotron 3.5 Super VL is 4.3× cheaper than Sonar on a typical 3:1 input/output mix. Nemotron 3.5 Super VL has the larger context window (1M vs 127K).
Nemotron 3.5 Super VLSonar
Input price / 1M$0.15$1.00
Output price / 1M$0.47$1.00
Blended price (3:1)$0.23$1.00
Cached input / 1M$0.02—
Context window1M127K
Max output944K114K
Intelligence index13.6—
Coding index——
Providers10—
Best provider uptime (30m)100.0%—
Vision inputYesYes
ReasoningYesNo
Tool callingYesNo
Released2026-10-152025-01-27
Retirement——

Cost calculator

FAQ

Which is cheaper, Nemotron 3.5 Super VL or Sonar?

Nemotron 3.5 Super VL is 4.3× cheaper than Sonar on a typical 3:1 input/output mix.

Nemotron 3.5 Super VL vs Sonar: which has a bigger context window?

Nemotron 3.5 Super VL: 1,048,576 tokens. Sonar: 127,072 tokens.

How much would 1 million requests cost?

At 2,000 input + 500 output tokens per request: Nemotron 3.5 Super VL ≈ $535.00, Sonar ≈ $2,500.

More comparisons