Granite Embedding English r2 vs Sonar Pro

This comparison does not currently contain a protocol-matched coding benchmark for both Granite Embedding English r2 and Sonar Pro, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.

Only Sonar Pro has a directly sourced input price: $3.00 per million tokens. Only Sonar Pro has a directly sourced output price: $15.00 per million tokens.

Sonar Pro has the larger sourced context window. Granite Embedding English r2 supports 8K and Sonar Pro supports 200K.

There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.

Granite Embedding English r2 is the only model in this pair currently marked as self-hostable. Granite Embedding English r2 is open weight; Sonar Pro is not marked open weight.

Granite Embedding English r2 is not documented with image input; Sonar Pro is not documented with image input. This reflects supported input modalities, not vision quality.

Neither has a larger sourced maximum output. Granite Embedding English r2 is — and Sonar Pro is —.

Granite Embedding English r2: none of these features are definitively sourced. Sonar Pro: none of these features are definitively sourced. Feature support does not establish relative quality.

Granite Embedding English r2 has 0 sourced provider routes; Sonar Pro has 2, so Sonar Pro has broader tracked availability.

There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.