Pairwise benchmark snapshot · Aug 17, 2026
In this head-to-head showdown, Doubao Lite 1.5 delivers higher overall intelligence and human-preferred responses (Elo 1,360 vs 1,320), while Codestral 25.01 leads in SWE-bench software engineering benchmarks (51.8%), while Doubao Lite 1.5 is more budget-friendly at $0.1/1M per 1M output tokens. Review the complete breakdown below to determine which model best fits your performance and budget requirements.
ByteDance high-speed lightweight model priced at sub-cent levels.
Speed, cost, and intelligence tradeoffs across the Mistral lineup
Δ 40 Arena Elo pts
180 tok/s
$0.1 / 1M output
Relative percentile scores computed across all active models in the benchmark catalog.
Each spoke is a skill. Farther from the center is better (0–100th percentile). Tap any dot to inspect details.
Category wins across reasoning intelligence, generation speed, and token cost.
Preference Elo, Coding proficiency, SWE-bench & LiveBench accuracy
Generation throughput and time to first token responsiveness
Cost per million tokens and max context window length
| Benchmark | Codestral 25.01 | Doubao Lite 1.5 | Advantage Delta |
|---|---|---|---|
| Preference Elo | Doubao Lite 1.5 (+40) | ||
| Coding Elo | Codestral 25.01 (+82) | ||
| LiveBench | — | ||
| SWE-bench | Codestral 25.01 (+5.8%) | ||
| GPQA Diamond | Doubao Lite 1.5 (+5.8%) | ||
| Time to first token | Doubao Lite 1.5 (+40 ms) | ||
| Output speed | 148 tok/s | 180 tok/s | Doubao Lite 1.5 (+32 tok/s) |
| Input price | $0.3/1M | $0.04/1M | Doubao Lite 1.5 (+$0.26/1M) |
| Output price | $0.9/1M | $0.1/1M | Doubao Lite 1.5 (+$0.8/1M) |
| Context window | Codestral 25.01 (+223k) |
Simulate monthly production API costs in USD (US Dollar).
Doubao Lite 1.5 is estimated to save $21.10/month ($253/year).
Higher SWE-bench (51.8%).
Lower output list price ($0.1/1M).
Lower TTFT (110 ms).
Larger window (256k).
Doubao Lite 1.5 is the side marked multimodal in the catalog.
| Target Workload | Recommended Pick | Evaluation Rationale |
|---|---|---|
| Repo / coding agents | Codestral 25.01 | Higher SWE-bench (51.8%). |
| High-volume chat | Doubao Lite 1.5 | Lower output list price ($0.1/1M). |
| Voice / low-latency UI | Doubao Lite 1.5 | Lower TTFT (110 ms). |
| Long-document RAG | Codestral 25.01 | Larger window (256k). |
| Screenshots / vision | Doubao Lite 1.5 | Doubao Lite 1.5 is the side marked multimodal in the catalog. |
No comments posted on this matchup yet. Be the first to share an evaluation note!
The real PNG is generated at /compare/codestral-25-vs-doubao-1-5-lite/opengraph-image for crawlers.
Verified head-to-head card
