Roboflow 100-VL
Models locate objects named in natural language across image collections from different domains. The benchmark scores predicted bounding boxes using average precision.
- Made by
- Roboflow
- Released
- May 2025
- Kind
- Detection and captioning
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Prediction
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- Based on
- A version of another benchmark
- In views
- Vision
- Last model added
- January 2025
- Scores refreshed
- November 20th, 2026
LW-DETRm tops the leaderboard with 59.8% Qwen2.5-VL-72B leads with 7.60%
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Qwen2.5-VL-72B, at 7.60%. This view shows only models we track, each at its best.
- 1 LW-DETRm Fully-Supervised 59.8%
- 2 LW-DETRs Fully-Supervised 57.4%
- 3 YOLOv11m Fully-Supervised 57%
- 4 YOLOv11s Fully-Supervised 57%
- 5 YOLOv8m Fully-Supervised 56.9%
- 6 YOLOv8s Fully-Supervised 56.5%
- 7 YOLOv11n Fully-Supervised 56.1%
- 8 YOLOv8n Fully-Supervised 55.4%
- 9 GroundingDINO Few-Shot (10 shots) 33.6%
- 10 Detic Few-Shot (10 shots), Federated Loss 22.8%
- 11 OWLv2 Zero-Shot, Class Names Only 13.6%
- 12 MQ-GLIP Few-Shot (10 shots), Class Names + Images 12.2%
- 13 MQ-GLIP-Text Zero-Shot, Class-Names Only 12.2%
- 14 Gemini 2.5 Pro (snapshot unspecified) Zero-Shot, Class Names Only 11.6%
- 15
Qwen2.5-VL-72B Zero-Shot, Instructions Only 7.60%
- 16 MQ-GLIP-Image Few-Shot (10 shots), Images Only 6.40%
Source: The Roboflow 100-VL paper
Source: The Roboflow 100-VL paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | LW-DETRm | 59.8% | Fully-Supervised | Official leaderboard | |
| 2 | LW-DETRs | 57.4% | Fully-Supervised | Official leaderboard | |
| 3 | YOLOv11m | 57% | Fully-Supervised | Official leaderboard | |
| 4 | YOLOv11s | 57% | Fully-Supervised | Official leaderboard | |
| 5 | YOLOv8m | 56.9% | Fully-Supervised | Official leaderboard | |
| 6 | YOLOv8s | 56.5% | Fully-Supervised | Official leaderboard | |
| 7 | YOLOv11n | 56.1% | Fully-Supervised | Official leaderboard | |
| 8 | YOLOv8n | 55.4% | Fully-Supervised | Official leaderboard | |
| 9 | GroundingDINO | 33.6% | Few-Shot (10 shots) | Official leaderboard | |
| 10 | Detic | 22.8% | Few-Shot (10 shots), Federated Loss | Official leaderboard | |
| 11 | OWLv2 | 13.6% | Zero-Shot, Class Names Only | Official leaderboard | |
| 12 | MQ-GLIP | 12.2% | Few-Shot (10 shots), Class Names + Images | Official leaderboard | |
| 13 | MQ-GLIP-Text | 12.2% | Zero-Shot, Class-Names Only | Official leaderboard | |
| 14 | Gemini 2.5 Pro (snapshot unspecified) | 11.6% | Zero-Shot, Class Names Only | Official leaderboard | |
| 15 | | 7.60% | Zero-Shot, Instructions Only | Jan 2025 | Official leaderboard |
| 16 | MQ-GLIP-Image | 6.40% | Few-Shot (10 shots), Images Only | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 7.60% | Zero-Shot, Instructions Only | Jan 2025 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The Roboflow 100-VL paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Roboflow 100-VL leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy.
Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy.
Qwen 3.8 Max 85.9%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 22 months ago, in January 2025, for Qwen2.5-VL-72B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (16 entries).
- Pass Models scored 1 model we track, plus 15 other entries on its leaderboard.
- Warning Newest score The newest came 22 months ago, in January 2025, for Qwen2.5-VL-72B.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 7.60%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 3 rows for models we track and 15 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.