agents benchmark
PinchBench Leaderboard
PinchBench evaluates coding agents on real-world agentic coding tasks across complex software engineering scenarios.
Updated Sep 5, 2026
Models6
Model coverage6
MetricScore
EvidenceB
PinchBench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelNV | Score90.00% | Percentile100.00% | Participants6 | EvidenceC | Evaluated |
| Rank02 | ModelNV | Score85.37% | Percentile80.00% | Participants6 | EvidenceC | Evaluated |
| Rank03 | ModelXI | Score81.20% | Percentile60.00% | Participants6 | EvidenceC | Evaluated |
| Rank04 | ModelXI | Score81.00% | Percentile40.00% | Participants6 | EvidenceC | Evaluated |
| Rank05 | ModelZA | Score80.70% | Percentile20.00% | Participants6 | EvidenceC | Evaluated |
| Rank06 | ModelLA | Score68.22% | Percentile0.00% | Participants6 | EvidenceC | Evaluated |
PinchBench Highlights
The leading models and scores on this benchmark.
PinchBench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for PinchBench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is PinchBench?
What PinchBench measures and how its scores work.
PinchBench is a benchmark for evaluating coding agents on real-world agentic coding tasks.
It measures Score as a ratio, including best-case and average performance across complex software engineering scenarios.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- PinchBench
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about PinchBench.
Which model scores highest on PinchBench?
Nemotron 3 Ultra (550B A55B) is currently ranked first with 90.00%.
What are the top three models on PinchBench?
The current leaders are Nemotron 3 Ultra (550B A55B) (90.00%), Nemotron 3.5 Lightning (30B A3B) (85.37%), and MiMo-V2-Omni (81.20%).
Which PinchBench model has the lowest official input price?
MiMo-V2-Omni has the lowest matched official input price at $0.14 input / $0.28 output per 1M tokens.
Which models are fastest among PinchBench results?
No matched runtime record is currently available.
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.