Skip to main content
Radar

Every change to the models you run, with its source and its date. Releases, price changes, retirements, API changes, and incidents.Every change to the models you run, with its source.

Follow model changes
BenchLM

CorpusQA 1M

A million-token CorpusQA long-context question-answering benchmark reported in DeepSeek-V4 model evaluations.

Data verified 33 confirmed releases in the last 30 daysSee provider release alerts

Benchmark score on CorpusQA 1M — September 15, 2026

We mirror the published score view for CorpusQA 1M. DeepSeek V4 Pro 0813 leads the public snapshot at 62.0%, followed by DeepSeek V4 Flash 0731 (60.5%) and DeepSeek V4 Flash (High) (59.3%). We do not use these results to rank models overall.

1Proprietary

DeepSeek V4 Pro 0813

DeepSeek

62.0%
Overall 66.57Context 1M
2Proprietary

DeepSeek V4 Flash 0731

DeepSeek

60.5%
Context 1M
3Proprietary

DeepSeek V4 Flash (High)

DeepSeek

59.3%
Context 1M

6 modelsReasoningCurrentDisplay onlyUpdated September 15, 2026

Benchmark score table (6 models)

Score
1
DeepSeek V4 Pro 0813DeepSeek · Closed
62.0%
2
DeepSeek V4 Flash 0731DeepSeek · Closed
60.5%
3
DeepSeek V4 Flash (High)DeepSeek · Closed
59.3%
4
DeepSeek V4 Pro (High)DeepSeek · Open weight
56.5%
5
DeepSeek V4 ProDeepSeek · Open weight
35.6%
6
DeepSeek V4 FlashDeepSeek · Closed
15.5%

The published CorpusQA 1M snapshot places DeepSeek V4 Pro 0813 first at 62.0%. The third row is 2.7 points behind. The broader top-10 range is 46.5 points, so the table still separates the published systems.

6 models have been evaluated on CorpusQA 1M. The benchmark falls in the Reasoning category. This category carries a 17% weight in BenchLM.ai's overall scoring system. CorpusQA 1M is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About CorpusQA 1M

Year

2026

Tasks

Million-token corpus question answering

Format

Long-context QA accuracy

Difficulty

Million-token long context

CorpusQA 1M is tracked as a display-only provider-table row for long-context DeepSeek-V4 comparisons.

DeepSeek-V4 Technical Report

BenchLM freshness & provenance

Version

CorpusQA 1M 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does CorpusQA 1M measure?

A million-token CorpusQA long-context question-answering benchmark reported in DeepSeek-V4 model evaluations.

Which model scores highest on CorpusQA 1M?

DeepSeek V4 Pro 0813 by DeepSeek currently leads with a score of 62.0% on CorpusQA 1M.

How many models are evaluated on CorpusQA 1M?

6 AI models have been evaluated on CorpusQA 1M on BenchLM.

Compare Top Models on CorpusQA 1M

DeepSeek V4 Pro 0813 vs DeepSeek V4 Flash 0731
Last updated: September 15, 2026 · BenchLM version CorpusQA 1M 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.