BenchmarksOfficeQA Pro (full image corpus)
OfficeQA Pro (full image corpus)
Document-grounded reasoning on original OfficeQA Pro using the full image-only PDF corpus.
Test definition
- Author / evaluator
- Reported by Moonshot AI
- Version / family
- OfficeQA Pro (full image corpus) / OfficeQA Pro
- Evidence available
- 5 results in 1 comparison groups
How the test is scored
Original OfficeQA Pro, with the entire PDF corpus supplied as images. This differs from preselected-document text-based OfficeQA reports and from Pro v2.
Comparison notes
Publisher-reported results, read 24 October 2026. Compare reported values only within the same source row; tool settings, subsets, versions, judges and metrics must not be combined.
A different task split, tool policy, judge or reasoning effort can change the result. Choose one comparison group below.
Results by evaluator
Select a publication and metric to compare the recorded field.
selfKimi K3.1 Pro: complete release evaluation · OfficeQA Pro
- 68.6%
- 67.4%
- 66.2%
- 62.4%
- 58.6%
Bars start at zero. 5 reported results in this comparison group; ties share a rank. Source ↗
All observations
Every numeric result, with its source and reported run conditions.
5 results
| Model | Result (%) | Source | Run |
|---|---|---|---|
| Kimi K3.1 Pro | 67.4OfficeQA Pro | self | DetailsOriginal OfficeQA Pro, with the entire PDF corpus supplied as images. This differs from preselected-document text-based OfficeQA reports and from Pro v2. Read 24 October 2026. All 47 source rows and six comparison columns are retained, with text-only HLE and both ZeroBench metrics split explicitly. Kimi uses max reasoning effort, temperature 1.0 and top-p 0.95 unless a row states otherwise; proprietary comparison columns are labelled max with fallback. Comparison figures include quoted external results and mixed harnesses. Missing cells remain null and do not enter ranks. Lower hallucination rates rank first; every other rank compares larger values within the same row. Terminal-Bench Science’s September DeepSeek score is attributed separately according to footnote 2. These publication claims do not populate independent boards. Highspeed has no independently identified results and inherits none of these scores.
|
| Claude Fable 5.1 | 68.6OfficeQA Pro | self | DetailsOriginal OfficeQA Pro, with the entire PDF corpus supplied as images. This differs from preselected-document text-based OfficeQA reports and from Pro v2. Read 24 October 2026. All 47 source rows and six comparison columns are retained, with text-only HLE and both ZeroBench metrics split explicitly. Kimi uses max reasoning effort, temperature 1.0 and top-p 0.95 unless a row states otherwise; proprietary comparison columns are labelled max with fallback. Comparison figures include quoted external results and mixed harnesses. Missing cells remain null and do not enter ranks. Lower hallucination rates rank first; every other rank compares larger values within the same row. Terminal-Bench Science’s September DeepSeek score is attributed separately according to footnote 2. These publication claims do not populate independent boards. Highspeed has no independently identified results and inherits none of these scores.
|
| GPT-6 Astra | 66.2OfficeQA Pro | self | DetailsOriginal OfficeQA Pro, with the entire PDF corpus supplied as images. This differs from preselected-document text-based OfficeQA reports and from Pro v2. Read 24 October 2026. All 47 source rows and six comparison columns are retained, with text-only HLE and both ZeroBench metrics split explicitly. Kimi uses max reasoning effort, temperature 1.0 and top-p 0.95 unless a row states otherwise; proprietary comparison columns are labelled max with fallback. Comparison figures include quoted external results and mixed harnesses. Missing cells remain null and do not enter ranks. Lower hallucination rates rank first; every other rank compares larger values within the same row. Terminal-Bench Science’s September DeepSeek score is attributed separately according to footnote 2. These publication claims do not populate independent boards. Highspeed has no independently identified results and inherits none of these scores.
|
| DeepSeek V4.1 Flash 1017 | 62.4OfficeQA Pro | self | DetailsOriginal OfficeQA Pro, with the entire PDF corpus supplied as images. This differs from preselected-document text-based OfficeQA reports and from Pro v2. Read 24 October 2026. All 47 source rows and six comparison columns are retained, with text-only HLE and both ZeroBench metrics split explicitly. Kimi uses max reasoning effort, temperature 1.0 and top-p 0.95 unless a row states otherwise; proprietary comparison columns are labelled max with fallback. Comparison figures include quoted external results and mixed harnesses. Missing cells remain null and do not enter ranks. Lower hallucination rates rank first; every other rank compares larger values within the same row. Terminal-Bench Science’s September DeepSeek score is attributed separately according to footnote 2. These publication claims do not populate independent boards. Highspeed has no independently identified results and inherits none of these scores.
|
| GLM 5.3 | 58.6OfficeQA Pro | self | DetailsOriginal OfficeQA Pro, with the entire PDF corpus supplied as images. This differs from preselected-document text-based OfficeQA reports and from Pro v2. Read 24 October 2026. All 47 source rows and six comparison columns are retained, with text-only HLE and both ZeroBench metrics split explicitly. Kimi uses max reasoning effort, temperature 1.0 and top-p 0.95 unless a row states otherwise; proprietary comparison columns are labelled max with fallback. Comparison figures include quoted external results and mixed harnesses. Missing cells remain null and do not enter ranks. Lower hallucination rates rank first; every other rank compares larger values within the same row. Terminal-Bench Science’s September DeepSeek score is attributed separately according to footnote 2. These publication claims do not populate independent boards. Highspeed has no independently identified results and inherits none of these scores.
|
Reproduction
Public execution instructions have not been verified in this catalogue. This status does not claim that the benchmark is closed or impossible to run.
Execution documentation ↗