reasoning benchmark
MBPP Leaderboard
MBPP (Mostly Basic Python Problems) is a benchmark of 974 crowd-sourced Python programming problems designed to be solvable by entry-level programmers, with each problem containing a task description, code solution, and 3 automated test cases.
Updated Sep 7, 2026
Models33
Model coverage33
MetricPass@1
EvidenceB
MBPP Ranking
Higher pass@1 ranks better on this benchmark.
Sort by
Rank | Model | Pass@1 | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelSA | Pass@192.70% | Percentile100.00% | Participants33 | EvidenceC | Evaluated |
| Rank02 | ModelNV | Pass@191.30% | Percentile96.88% | Participants33 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Pass@190.20% | Percentile93.75% | Participants33 | EvidenceC | Evaluated |
| Rank04 | ModelOP | Pass@189.11% | Percentile90.63% | Participants33 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Pass@188.20% | Percentile87.50% | Participants33 | EvidenceC | Evaluated |
| Rank06 | ModelNV | Pass@184.60% | Percentile84.38% | Participants33 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Pass@184.00% | Percentile81.25% | Participants33 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Pass@184.00% | Percentile78.13% | Participants33 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Pass@183.50% | Percentile75.00% | Participants33 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Pass@182.00% | Percentile71.88% | Participants33 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Pass@181.40% | Percentile68.75% | Participants33 | EvidenceC | Evaluated |
| Rank12 | ModelMI | Pass@180.80% | Percentile65.63% | Participants33 | EvidenceC | Evaluated |
| Rank13 | ModelAC | Pass@180.20% | Percentile62.50% | Participants33 | EvidenceC | Evaluated |
| Rank14 | ModelAC | Pass@179.20% | Percentile59.38% | Participants33 | EvidenceC | Evaluated |
| Rank15 | ModelMA | Pass@178.20% | Percentile56.25% | Participants33 | EvidenceC | Evaluated |
| Rank16 | ModelME | Pass@177.60% | Percentile53.13% | Participants33 | EvidenceC | Evaluated |
| Rank17 | ModelGO | Pass@176.00% | Percentile50.00% | Participants33 | EvidenceC | Evaluated |
| Rank18 | ModelMA | Pass@174.71% | Percentile46.88% | Participants33 | EvidenceC | Evaluated |
| Rank19 | ModelGO | Pass@174.40% | Percentile43.75% | Participants33 | EvidenceC | Evaluated |
| Rank20 | ModelAC | Pass@173.20% | Percentile40.63% | Participants33 | EvidenceC | Evaluated |
| Rank21 | ModelGO | Pass@173.00% | Percentile37.50% | Participants33 | EvidenceC | Evaluated |
| Rank22 | ModelMA | Pass@169.64% | Percentile34.38% | Participants33 | EvidenceC | Evaluated |
| Rank23 | ModelMI | Pass@169.60% | Percentile31.25% | Participants33 | EvidenceC | Evaluated |
| Rank24 | ModelME | Pass@167.80% | Percentile28.13% | Participants33 | EvidenceC | Evaluated |
| Rank25 | ModelAC | Pass@167.20% | Percentile25.00% | Participants33 | EvidenceC | Evaluated |
| Rank26 | ModelGO | Pass@163.60% | Percentile21.88% | Participants33 | EvidenceC | Evaluated |
| Rank27 | ModelGO | Pass@163.60% | Percentile18.75% | Participants33 | EvidenceC | Evaluated |
| Rank28 | ModelGO | Pass@163.20% | Percentile15.63% | Participants33 | EvidenceC | Evaluated |
| Rank29 | ModelGO | Pass@162.60% | Percentile12.50% | Participants33 | EvidenceC | Evaluated |
| Rank30 | ModelGO | Pass@156.60% | Percentile9.38% | Participants33 | EvidenceC | Evaluated |
MBPP Highlights
The leading models and scores on this benchmark.
MBPP Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MBPP
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is MBPP?
What MBPP measures and how its scores work.
MBPP (Mostly Basic Python Problems) is a benchmark covering Python programming fundamentals and standard library functionality.
It reports Pass@1 as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MBPP
- Modality
- text
- Primary category
- reasoning
- 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 MBPP.
Which model scores highest on MBPP?
Sarvam-30B is currently ranked first with 92.70%.
What are the top three models on MBPP?
The current leaders are Sarvam-30B (92.70%), Llama-3.3 Nemotron Super 49B v1 (91.30%), and Qwen2.5-Coder 32B Instruct (90.20%).
Which MBPP model has the lowest official input price?
Qwen2.5-Omni-7B has the lowest matched official input price at $0.10 input / $0.40 output per 1M tokens.
Which models are fastest among MBPP results?
The fastest matched records are Llama 4 Scout (776.10 tok/s via Groq), Llama 4 Maverick (307.30 tok/s via Groq), and Qwen2.5-Coder 32B Instruct (44.00 tok/s via DeepInfra).
Does the highest pass@1 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.