Knowledge · Benchmark profile
A Chinese-language academic and professional benchmark spanning humanities, social science, STEM, and applied subjects.
Data verified 21 Jul 2026 · Methodology 1.6.0
Benchmark score on C-Eval
Qwen3.6 Plus leads at 93.3%, followed by DeepSeek V4 Pro Base (93.1%) and Qwen3.5 397B (93%).
Visual analysis
Switch between model placement, score distribution and descriptive provider averages. Every view uses the same sourced leaderboard.
7 ranked models · higher is better
| Model | Rank | Provider | Score |
|---|---|---|---|
| Qwen3.6 Plus | #1 | Alibaba Cloud | 93.3% |
| DeepSeek V4 Pro Base | #2 | DeepSeek | 93.1% |
| Qwen3.5 397B | #3 | Alibaba Cloud | 93% |
| Claude Opus 4.5 | #4 | Anthropic | 92.2% |
| DeepSeek V4 Flash Base | #5 | DeepSeek | 92.1% |
| Qwen3.6 27B | #6 | Alibaba Cloud | 91.4% |
| Qwen3.6-35B-A3B | #7 | Alibaba Cloud | 90% |
One best score per model · higher is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | Qwen3.6 Plus qwen3-6-plus | Alibaba Cloud | closed | Reference only | 93.3% |
| #2 | DeepSeek V4 Pro Base deepseek-v4-pro-base | DeepSeek | open | Estimated reference | 93.1% |
| #3 | Qwen3.5 397B qwen3.5-397b-a17b | Alibaba Cloud | open | Estimated reference | 93% |
| #4 | Claude Opus 4.5 claude-opus-4-5 | Anthropic | closed | Reference only | 92.2% |
| #5 | DeepSeek V4 Flash Base deepseek-v4-flash-base | DeepSeek | open | Estimated reference | 92.1% |
| #6 | Qwen3.6 27B qwen3.6-27b | Alibaba Cloud | open | Estimated reference | 91.4% |
| #7 | Qwen3.6-35B-A3B qwen3-6-35b-a3b | Alibaba Cloud | open | Estimated reference | 90% |
The top of this snapshot is led by Qwen3.6 Plus at 93.3%; third place is 0.30 points behind. The top-7 spread is 3.3 points.
About C-Eval
A Chinese-language academic and professional benchmark spanning humanities, social science, STEM, and applied subjects. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
A Chinese-language academic and professional benchmark spanning humanities, social science, STEM, and applied subjects.
Qwen3.6 Plus by Alibaba Cloud currently leads with 93.3%.
7 models in the LuminaBench cohort have a qualifying score on this benchmark.
No. This benchmark is display-only and does not enter the overall Lumina composite.
Related
2026 · 10 results · Reference
Artificial Analysis MMLU-Pro2026 · 4 results · Reference
Massive Multitask Language Understanding2020 · 8 results · Reference
Graduate-Level Google-Proof Q&A2023 · 70 results · Reference
SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines2025 · 19 results · Reference
Massive Multitask Language Understanding Professional2024 · 42 results · Reference
