Models / DeepSeek
DeepSeek V4 Pro 0813
by DeepSeek ·
deepseek/deepseek-v4-pro-0813DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.
reasoningtool-usereleased 2026-08-12
Legal score · raw
-
not yet benchmarked
Context
1.05M
max output 944K
Input
$1.12
per 1M tokens
Output
$3.37
per 1M tokens
Suite cost
-
run the suite to see
Benchmark results
| Task | Category | Raw | Juiced | Correct | Latency | Cost | Ran |
|---|---|---|---|---|---|---|---|
| Hearsay Identification | Evidence | - | - | - | - | - | - |
| Bluebook Citation Format | Research & Writing | - | - | - | - | - | - |
| Federal Civil Procedure | Procedure | - | - | - | - | - | - |
| Limitations Arithmetic | Procedure | - | - | - | - | - | - |
| Contract Clause Classification | Contracts | - | - | - | - | - | - |
| Citation Hallucination Resistance | Reliability | - | - | - | - | - | - |
Measured by DocketBuster
These numbers come from DocketBuster's own legal battery, not from DocketRouter's suite. Latest run per metric, with n and a 95% Wilson interval where the source reports one. See docketbuster.com/benchmarks.
| Metric | Value | n | Interval | Measured |
|---|---|---|---|---|
| Statute pinpoint, exact section (no retrieval) | 23.3% (70/300) | 300 | 95% CI 18.9% to 28.4% | 2026-08-22 |
| Statute pinpoint, exact section (with DocketBuster retrieval) | 13.4% (41/306) | 306 | 95% CI 10.0% to 17.7% | 2026-08-25 |
| Say-nothing rate (declines to bluff when the answer is not in the record) | 100.0% | 264 | 95% CI 98.6% to 100.0% | 2026-08-22 |
| Abstained on statute pinpoint | 74.5% (228/306) | 306 | count, no interval reported | 2026-08-25 |
| Coaching quality (GW-14x, 0 to 8) | 7.48 / 8 | - | rubric mean, no interval reported | 2026-08-23 |
Source files: hard-llm-ortier-dsv4pro.json, gw14x-ortier-dsv4pro.json, hard-llm-level-dsv4pro-statute_rag.json. Raw model name in source: deepseek/deepseek-v4-pro-0813.