Pairwise benchmark snapshot · Aug 17, 2026
This page compares Claude 3.7 Sonnet (Anthropic) and Doubao Pro 1.5 (ByteDance) using the latest snapshots we have as of Aug 17, 2026. On preference Elo, Doubao Pro 1.5 is ahead by 108 points (1,470 vs 1,362; lmarena (Aug 17, 2026)). On SWE-bench coding, Claude 3.7 Sonnet resolves 70.3% versus 62.1% (swebench (Aug 17, 2026)). Doubao Pro 1.5 streams faster (105 tok/s) while Doubao Pro 1.5 is the cheaper output token ($0.28/1M). Claude 3.7 Sonnet has the larger context window (200k tokens). Every cell below is a dated snapshot from a named public source — we do not invent missing scores, and we do not run SWE-bench ourselves.
Anthropic
Claude 3.7 SonnetHybrid-reasoning Sonnet from 2025. Kept for historical compare pages.
ByteDance
Doubao Pro 1.5ByteDance flagship enterprise model with ultra-low token cost and 128k context.
thinking_budgetReasoning models scale test-time compute by generating hidden chain-of-thought (CoT) tokens. These tokens are billed at standard output rates and directly expand latency (TTFT) in exchange for higher GPQA Diamond and SWE-bench Verified problem resolve rates.
Official evaluation benchmark budget for SWE-bench and LiveBench.
"thinking": {
"type": "enabled",
"budget_tokens": 16384
}"reasoning": {
"effort": "medium",
"max_tokens": 3200
}Instant pairwise benchmark recalculation across 400+ LLMs
Category wins across intelligence, speed, and pricing efficiency.
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
Each spoke is a skill. Farther from the center is better (0–100th percentile). Tap any dot to inspect details.
108 pts advantage
20 tok/s faster
$14.72/1M cheaper
Simulate monthly production API costs in USD (US Dollar).
Doubao Pro 1.5 is estimated to save $321.95/month ($3,863/year).
Higher SWE-bench (70.3%).
Lower output list price ($0.28/1M).
Lower TTFT (190 ms).
Larger window (200k).
Both accept images. Defaulting to the higher-Elo side (Doubao Pro 1.5).
| Target Workload | Recommended Pick | Evaluation Rationale |
|---|---|---|
| Repo / coding agents | Claude 3.7 Sonnet | Higher SWE-bench (70.3%). |
| High-volume chat | Doubao Pro 1.5 | Lower output list price ($0.28/1M). |
| Voice / low-latency UI | Doubao Pro 1.5 | Lower TTFT (190 ms). |
| Long-document RAG | Claude 3.7 Sonnet | Larger window (200k). |
| Screenshots / vision | Doubao Pro 1.5 | Both accept images. Defaulting to the higher-Elo side (Doubao Pro 1.5). |
No comments posted on this matchup yet. Be the first to share an evaluation note!
The real PNG is generated at /compare/claude-3-7-sonnet-vs-doubao-1-5-pro-128k/opengraph-image for crawlers.
Verified head-to-head card