jamba-large-1.7 vs sonar-pro
jamba-large-1.7 is 1.5× cheaper on input than sonar-pro.
| jamba-large-1.7 | sonar-pro | |
|---|---|---|
| Input / 1M | $2.00 | $3.00 |
| Cached input / 1M | — | — |
| Output / 1M | $8.00 | $15.00 |
| Context window | 256K | 200K |
| Max output | 256K | 8K |
| Provider | AI21 | Perplexity |
| Vision | No | No |
| Tool calling | No | No |
| Token count | estimated | estimated |
Capability for the money
Independent measurement · Artificial Analysis · read 2026-08-11Not the same tier: sonar-pro leads jamba-large-1.7 by 18.8 points on hard reasoning (GPQA Diamond), for 1.7× the price.
Coding
writing and fixing code
LiveCodeBenchsonar-pro +9.4
Hard reasoning
graduate-level science and multi-step analysis
GPQA Diamondsonar-pro +18.8
Averaged over the benchmarks both models sat, jamba-large-1.7 returns 8.2 points of pass rate per dollar of blended price against 7.1 — the ratio that decides whether the dearer model earns its rate.
Each figure is a pass rate: the share of that test set the model answered correctly, measured by a third party rather than self-reported by the lab. Rows are different tests of different difficulty, so compare a model against the other model, not one row against the next. No published following instructions or tool use score for one of these models, so those lenses are omitted. And a benchmark is not your workload — it narrows the shortlist, it does not pick for you.
Which to pick
- Cheaper input: jamba-large-1.7
- Cheaper output: jamba-large-1.7
- Larger context: jamba-large-1.7
- Better at coding: sonar-pro (18.1% vs 27.5% on LiveCodeBench)
- Better at hard reasoning: sonar-pro (39.0% vs 57.8% on GPQA Diamond)
List price is not production cost. Output length, cache hit rate, retries and task quality move the real number more than the headline rate does — price your own prompt on the token ledger before deciding.
Related comparisons
jamba-large-1.7 detail · sonar-pro detail · All AI21 pricing · All comparisons