reasoning benchmark
BigCodeBench-Hard Leaderboard
BigCodeBench-Hard is a subset of 148 challenging programming tasks from BigCodeBench covering complex, real-world programming problems across 7 domains and 139 libraries.
Updated Sep 5, 2026
Models1
Model coverage1
MetricScore
EvidenceB
BigCodeBench-Hard Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score27.00% | Percentile100.00% | Participants1 | EvidenceC | Evaluated |
BigCodeBench-Hard Highlights
The leading models and scores on this benchmark.
The Top AI Models for BigCodeBench-Hard
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis bigcodebench-hard AI model leaderboard uses descending score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ACAlibaba Cloud / Qwen Team
- Score
- 27.00%
- Speed
- Up to 44.00 tok/s via DeepInfra
Strengths
- Ranks #1 of 1 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures BigCodeBench-Hard, not total model capability
Selection summary
Best AI Models for BigCodeBench-Hard
Qwen2.5-Coder 32B Instruct currently leads BigCodeBench-Hard with 27.00%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price, and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is BigCodeBench-Hard?
What BigCodeBench-Hard measures and how its scores work.
BigCodeBench-Hard is a reasoning benchmark comprising 148 challenging programming tasks from BigCodeBench, with an average of 2.8 libraries per task.
It measures large language models' ability to solve complex, real-world programming problems, use diverse function calls from multiple libraries, apply compositional reasoning, and implement complex instructions; its metric is Score, reported as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- BigCodeBench-Hard
- 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 BigCodeBench-Hard.
Which model scores highest on BigCodeBench-Hard?
Qwen2.5-Coder 32B Instruct is currently ranked first with 27.00%.
What are the top three models on BigCodeBench-Hard?
The current leaders are Qwen2.5-Coder 32B Instruct (27.00%).
Which BigCodeBench-Hard model has the lowest official input price?
No matched official input price is currently available.
Which models are fastest among BigCodeBench-Hard results?
The fastest matched records are Qwen2.5-Coder 32B Instruct (44.00 tok/s via DeepInfra).
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.