QuanBench+ - Qiskit: leaderboard

Quantum code generation in Qiskit on QuanBench+ (2026): 42 tasks on quantum algorithms, gate decomposition and state preparation, aligned across Qiskit, Cirq and PennyLane and checked by executable functional tests (probabilistic outputs accepted by KL divergence). Pass@1 with greedy decoding, in percent; higher is better.

Metric: Pass@1 (%): greedy decoding at temperature 0, one Qiskit program per task, on the 42 framework-aligned QuanBench+ tasks (quantum algorithms, gate decomposition, state preparation), graded by executable functional tests with KL-divergence acceptance for probabilistic outputs; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 12 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro59.5#146
2GPT-5.157.1#248
3GPT-4.150#506
4DeepSeek R150#484
5Kimi K2 (Thinking)47.6
6Claude 3.7 Sonnet45.2#455
7QuanBench+ DeepSeek-Chat (checkpoint unspecified)45.2
8GLM-4.742.9#383
9Gemini 2.5 Flash40.5#466
10Llama 4 Maverick38.1#1113
11MiniMax-M2.128.6#582
12Qwen 2.5 7B Instruct16.7#2732

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=quanbench-plus-qiskit · How It Works · Data refreshed daily, snapshot 2026-11-22.