Sonar vs Codestral vs Qwen Plus Character (Japanese)
Codestral comes out ahead, 48 to 36 and 30 on our weighted score, and it is the cheaper option too.
PerplexitySonar
30/100- ECI—
- Price$1.00 / $1.00
- Context128K
- Our pick
Mistral AICodestral
48/100- ECI—
- Price$0.30 / $0.90
- Context256K
Alibaba (Qwen)Qwen Plus Character (Japanese)
36/100- ECI—
- Price$0.50 / $1.40
- Context8K
Codestral is our pick
Codestral is the better all-round choice, scoring 48/100 against Qwen Plus Character (Japanese) (36) and Sonar (30). It leads on price and context window. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceCodestralCodestral $0.45 · Qwen Plus Character (Japanese) $0.725 · Sonar $1.00 per 1M tokens (3:1 blend)
- Longest contextCodestralCodestral 256,000 · Sonar 128,000 · Qwen Plus Character (Japanese) 8,192 tokens
- Widest inputsSame inputsSonar: Text · Codestral: Text · Qwen Plus Character (Japanese): Text
- Self-hostingCodestralPublishes downloadable weights
| Measure | Weight | Sonar | Codestral | Qwen Plus Character (Japanese) |
|---|---|---|---|---|
| Price | 50% | 50 | 66 | 57 |
| Inputs & features | 30% | 0 | 25 | 25 |
| Context window | 20% | 24 | 36 | 0 |
| Overall | 100% | 30/100 | 48/100 | 36/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | SonarPerplexity | CodestralMistral AI | Qwen Plus Character (Japanese)Alibaba (Qwen) |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $1.00 | $0.30 (best) | $0.50 |
| Output | $1.00 | $0.90 (best) | $1.40 |
| Cached input | — | $0.03 | — |
| Blended (3:1) | $1.00 | $0.45 (best) | $0.725 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official Perplexity API | Official Mistral API | Official Alibaba API |
| Limits | |||
| Context window | 128,000 tokens | 256,000 tokens (best) | 8,192 tokens |
| Max output | 4,096 tokens (best) | 4,096 tokens (best) | 512 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | No | No | No |
| Tool calling | No | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Proprietary | Open | Proprietary |
| API model ID | sonar | codestral-latest | qwen-plus-character-ja |
| API providers | 6 (best) | 3 | 1 |
| Released | Jan 1, 2024 | May 29, 2024 | Jan 2024 |
| Knowledge cutoff | Sep 1, 2025 | Oct 2024 | Apr 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Sonar$12.00
- Codestral$4.80
- Qwen Plus Character (Japanese)$7.80
Which should you choose?
Which is better: Sonar, Codestral or Qwen Plus Character (Japanese)?
Codestral is the better all-round choice, scoring 48/100 against Qwen Plus Character (Japanese) (36) and Sonar (30). It leads on price and context window. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Sonar, Codestral or Qwen Plus Character (Japanese)?
Codestral is cheaper at $0.30 input / $0.90 output per million tokens (official Mistral API price). Qwen Plus Character (Japanese) costs $0.50 input / $1.40 output per million tokens (official Alibaba API price); Sonar costs $1.00 input / $1.00 output per million tokens (official Perplexity API price). At a typical mix of three input tokens to one output token, that is $0.45 per million tokens for Codestral versus $0.725 for Qwen Plus Character (Japanese) (1.6× as much) and $1.00 for Sonar (2.2× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Sonar has not been scored yet, Codestral has not been scored yet and Qwen Plus Character (Japanese) has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Sonar, Codestral and Qwen Plus Character (Japanese) yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Sonar does not support tool calling, which most coding agents need.
Which has the bigger context window?
Codestral has the largest context window at 256,000 tokens, against 128,000 for Sonar and 8,192 for Qwen Plus Character (Japanese). Maximum output per response: Sonar up to 4,096, Codestral up to 4,096, Qwen Plus Character (Japanese) up to 512 tokens.
Which can read images, PDFs, audio or video?
Sonar accepts text; Codestral accepts text; Qwen Plus Character (Japanese) accepts text. They handle the same number of input types.
Are any of these open source?
Codestral publishes its weights and can be self-hosted; Sonar and Qwen Plus Character (Japanese) is proprietary.
Which is newer?
Codestral is the newest, released May 29, 2024. Sonar came out Jan 1, 2024; Qwen Plus Character (Japanese) came out Jan 2024. Knowledge cutoff: Sonar Sep 1, 2025, Codestral Oct 2024, Qwen Plus Character (Japanese) Apr 2024.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
