CharXiv (RQ)
CharXiv (RQ) - reasoning questions over arXiv figures and tables.
Category
Document Understanding
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for CharXiv (RQ)
| Model | Score | Date | Verified |
|---|---|---|---|
| Qwen3.6-35B-A3B |
47.43%
|
16.04.2026 | Verified |
| Gemma4-31B-it |
7.51%
|
16.04.2026 | Unverified |
| Qwen3.5-35B-A3B |
45.45%
|
16.04.2026 | Unverified |
| Gemma4-26B-A4B |
11.86%
|
16.04.2026 | Unverified |
| Claude-Sonnet-4.5 |
4.74%
|
16.04.2026 | Unverified |
| Qwen3.6-27B |
49.01%
|
22.04.2026 | Verified |
| Claude 4.5 Opus |
9.88%
|
22.04.2026 | Unverified |
| Qwen3.5-27B |
53.36%
|
16.04.2026 | Unverified |
| Qwen3.8-27B |
69.96%
|
18.08.2026 | Verified |
| Qwen3.7-Plus |
78.26%
|
18.08.2026 | Unverified |
| Muse Glimmer-30B |
50.59%
|
18.08.2026 | Unverified |
| Opus4.6 Max |
66.00
|
18.08.2026 | Unverified |
| Qwen3.5-4B |
18.97%
|
01.02.2026 | Verified |
| Muse-Glimmer-30B |
50.59%
|
01.08.2026 | Verified |
| Kimi K3 |
100.00%
|
18.08.2026 | Verified |
| Qwen3.5-122B-A10B |
44.27%
|
01.02.2026 | Verified |
| GPT-5-mini 2025-08-07 |
10.28%
|
01.02.2026 | Unverified |
| Qwen3-235B-A22B |
0.40%
|
01.02.2026 | Unverified |
| Qwen3-VL-235B-A22B |
0.40%
|
01.02.2026 | Verified |
| Kimi K2.6 |
56.92%
|
23.08.2026 | Verified |
| GPT-5.4 |
66.40%
|
23.08.2026 | Verified |
| Claude Opus 4.6 |
12.25%
|
23.08.2026 | Verified |
| Gemini 3.1 Pro |
56.13%
|
23.08.2026 | Verified |
| Kimi K2.5 |
45.45%
|
23.08.2026 | Verified |
| Qwen3.8-Flash-Next |
73.52%
|
26.08.2026 | Verified |
| Claude-Opus-4.6 (Max) |
66.00
|
26.08.2026 | Unverified |
| Qwen3.5-397B-A17B |
58.50%
|
22.04.2026 | Unverified |