Recent Papers / arXiv:pwc-82956
DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
Authors pending
Tasks
editResults
10 results reproduced from this paper.
Sorted instantly in-page
Results
8
SOTA rows
2
Models
2
Datasets
0
| # | Model | Vendor | Benchmark | Value | SOTA | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | DeepSeek-V4-Pro Max | DeepSeek | GPQA Diamond | 90.1% | — | source ↗ | |
| 02 | DeepSeek-V4-Flash Max | DeepSeek | GPQA Diamond | 88.1% | — | source ↗ | |
| 03 | DeepSeek-V4-Pro Max | DeepSeek | BrowseComp | 83.4% | #1 | — | source ↗ |
| 04 | DeepSeek-V4-Pro Max | DeepSeek | SWE-Bench Verified | 80.6% | #1 | — | source ↗ |
| 05 | DeepSeek-V4-Flash Max | DeepSeek | SWE-Bench Verified | 79.0% | — | source ↗ | |
| 06 | DeepSeek-V4-Flash Max | DeepSeek | BrowseComp | 73.2% | — | source ↗ | |
| 07 | DeepSeek-V4-Pro Max | DeepSeek | HLE | 37.7% | — | source ↗ | |
| 08 | DeepSeek-V4-Flash Max | DeepSeek | HLE | 34.8% | — | source ↗ |
CodeSOTA extraction
Benchmark evidence
Link this paper to benchmark rows, datasets, model cards, and reproduced results as evidence is extracted.