Model comparison
DeepSeek vs Qwen API
DeepSeek and Qwen are both strong Chinese LLM families, but GPTokens benchmark notes point to different starting points depending on tool calling, structured output, and agent risk.
Route shortlist
Benchmark comparison
| Model | Total /100 | Extraction | Instructions | Math | Tools | Fully correct | Tokens | Cost (USD) | CPS ($/100 points) |
|---|---|---|---|---|---|---|---|---|---|
DeepSeek v4 Prodeepseek-v4-pro | 99.00 | 98.33 | 98.67 | 99.00 | 100.00 | 57/60 | 116,757 | $0.145106 | $0.146572 |
Qwen 3.7 Plusqwen3.7-plus | 92.72 | 99.56 | 100.00 | 72.00 | 99.33 | 52/60 | 122,334 | $0.116596 | $0.125748 |
BenchLocal V3.2 · Original 27 models: 2 October 2026; three full-suite supplements: 15 October 2026 · 60 tasks per model. Total = mean of the four category scores. The original models retain 45 V3 tasks and 15 revised tool tasks; the three additional model IDs each completed all 60 identical tasks. Near-ceiling scores do not establish general capability; V3 scores are not directly comparable. CPS = Cost per Score; shown as USD per 100 score points: retained execution cost ÷ unrounded total score × 100. Free models are shown at $0. Paid-model costs are historical charges, excluding failed requests and discarded retests.
Decision guide
Recommended route
Start with Qwen 3.7 Plus for structured output and DeepSeek v4 Pro for tool-heavy agent experiments.
Best fit
Qwen 3.7 Plus scored 90 overall in the earlier pre-V3 evaluation and is especially strong for structured output and data extraction. DeepSeek v4 Pro scored 89 and is a strong tool-calling and agent candidate.
API access
Both can be tested through GPTokens with the same OpenAI-compatible base URL, API key pattern, and /v1/models discovery workflow.
Related pages