Model comparison
Qwen Turbo vs Sonar
Sonar is the stronger model overall, scoring 37.8 to 26.8 on the Noometry Index. Qwen Turbo costs 11× less per token, which makes it the better buy when Sonar's lead doesn't matter for your workload.
Last verified . 0 shared benchmarks.
Alibaba (Qwen)
PerplexitySummary
- The widest gap is in math, where Sonar leads 31.3 to 14.5.
- Qwen Turbo is cheaper at $0.05 / $0.20 per million input/output tokens, against $1 / $1 for Sonar.
- Qwen Turbo accepts more context: 1M tokens versus 128K.
Side by side
| Qwen Turbo | Sonar | |
|---|---|---|
| Provider | Alibaba (Qwen) | Perplexity |
| Noometry Index | 26.8 | 37.8 |
| Released | 2024-11-01 | 2024-01-01 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 128K |
| Max output | 16K | 4K |
| Input $ / M tokens | $0.05 | $1 |
| Output $ / M tokens | $0.20 | $1 |
| Results tracked | 3 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Qwen Turbo: —, Sonar: 35.1 (#217)
| Benchmark | Qwen Turbo | Sonar |
|---|---|---|
| LiveBench Coding | — | 35.1% |
Reasoning Not comparable
Qwen Turbo: —, Sonar: 20.3 (#221)
| Benchmark | Qwen Turbo | Sonar |
|---|---|---|
| LiveBench Reasoning | — | 46.3% |
| LiveBench Data Analysis | — | 37.9% |
| LiveBench | — | 46.9% |
Math Sonar leads
Qwen Turbo: 14.5 (#291), Sonar: 31.3 (#197)
| Benchmark | Qwen Turbo | Sonar |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.1% | — |
| LiveBench Math | — | 41.6% |
| MATH Level 5 | 56.2% | — |
Knowledge Not comparable
Qwen Turbo: 22.2 (#267), Sonar: —
| Benchmark | Qwen Turbo | Sonar |
|---|---|---|
| GPQA Diamond | 41.8% | — |
Instruction Following Not comparable
Qwen Turbo: —, Sonar: 71.5 (#146)
| Benchmark | Qwen Turbo | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
Writing & Preference Not comparable
Qwen Turbo: —, Sonar: 52.6 (#166)
| Benchmark | Qwen Turbo | Sonar |
|---|---|---|
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Qwen Turbo better than Sonar?
Sonar is the stronger model overall, scoring 37.8 to 26.8 on the Noometry Index. Qwen Turbo costs 11× less per token, which makes it the better buy when Sonar's lead doesn't matter for your workload.
Which is cheaper, Qwen Turbo or Sonar?
Qwen Turbo is cheaper. It lists at $0.05 per million input tokens and $0.20 per million output tokens; Sonar lists at $1 and $1.
Which has the bigger context window?
Qwen Turbo does, with 1M tokens against 128K.
How many benchmarks do Qwen Turbo and Sonar share?
0 benchmarks have published results for both models. Qwen Turbo has 3 scored results on Noometry and Sonar has 7.