PowerCodeBench
Models translate operators' requests into Python programs for pandapower. The evaluator runs each program and compares its numerical or Boolean output with the reference answer.
Grid Studio scored 352 of the 2,000 tasks. It restored missing specifications from the source records for 156 of those questions.
- Made by
- University of Exeter
- Released
- May 2026
- Kind
- Scientific code
- Level
- University
- Scored
- Checked automatically
- Answers
- Code
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Environment
- Code sandbox
- Refresh
- One fixed set
- Run by
- Anyone
- Occupations
- Production
- Strongest model tested
- GPT-5.4 Mini #74 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
Llama 3.1-405B leads with 68.7% among models released by April 2026 Llama 3.1-405B leads with 68.7% among models released by April 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 68.7
Llama 3.1-405B
- 67.3
Qwen3-Coder-480B-A35B
- 63.8 Claude Haiku 4.5
- 62.1 GPT-5.4 Mini
- 60.9
Qwen3-Coder-Next
- 60.3 gpt-oss-120b
- 57.5
Llama 3.1-70B
- 55.9
DeepSeek-V4-Flash
- 54.3
Qwen2.5-Coder-32B-Instruct
- 52.0
Gemini 2.5 Flash
- 51.1 Qwen2.5-Coder-14B-Instruct
- 36.5
Llama 3.1-8B
- 32.8 Qwen2.5-Coder-7B-Instruct
- 9.90 Qwen2.5-Coder-1.5B-Instruct
- 0.60 Qwen2.5-Coder-0.5B-Instruct
- 1
Llama 3.1-405B C_FDRS 68.7%
- 2
Qwen3-Coder-480B-A35B X_FDRS 67.3%
- 3 Claude Haiku 4.5 C_FDRS 63.8%
- 4 GPT-5.4 Mini C_FDRS 62.1%
- 5
Qwen3-Coder-Next X_FDRS 60.9%
- 6 gpt-oss-120b C_FDRS 60.3%
- 7
Llama 3.1-70B C_FDRS 57.5%
- 8
DeepSeek-V4-Flash C_FDRS 55.9%
- 9
Qwen2.5-Coder-32B-Instruct X_FD 54.3%
- 10
Gemini 2.5 Flash Sep 2025, C_FDRS 52%
- 11 Qwen2.5-Coder-14B-Instruct X_FDRS 51.1%
- 12
Llama 3.1-8B X_FDRS 36.5%
- 13 Qwen2.5-Coder-7B-Instruct C_FDRS 32.8%
- 14 Qwen2.5-Coder-1.5B-Instruct C_FX 9.90%
- 15 Qwen2.5-Coder-0.5B-Instruct X_FD 0.60%
Source: Official PowerCodeBench leaderboard
- 1
Llama 3.1-405B C_FDRS 68.7%
- 2
Qwen3-Coder-480B-A35B X_FDRS 67.3%
- 3 Claude Haiku 4.5 C_FDRS 63.8%
- 4 GPT-5.4 Mini C_FDRS 62.1%
- 5
Qwen3-Coder-Next X_FDRS 60.9%
- 6 gpt-oss-120b C_FDRS 60.3%
- 7
Llama 3.1-70B C_FDRS 57.5%
- 8
DeepSeek-V4-Flash C_FDRS 55.9%
- 9
Qwen2.5-Coder-32B-Instruct X_FD 54.3%
- 10
Gemini 2.5 Flash Sep 2025, C_FDRS 52%
- 11 Qwen2.5-Coder-14B-Instruct X_FDRS 51.1%
- 12
Llama 3.1-8B X_FDRS 36.5%
Source: Official PowerCodeBench leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 68.7% | C_FDRS | Jul 2024 | Official leaderboard |
| 2 | | 67.3% | X_FDRS | Jul 2025 | Official leaderboard |
| 3 | Claude Haiku 4.5 | 63.8% | C_FDRS | Oct 2025 | Official leaderboard |
| 4 | GPT-5.4 Mini | 62.1% | C_FDRS | Mar 2026 | Official leaderboard |
| 5 | | 60.9% | X_FDRS | Feb 2026 | Official leaderboard |
| 6 | gpt-oss-120b | 60.3% | C_FDRS | Aug 2025 | Official leaderboard |
| 7 | | 57.5% | C_FDRS | Jul 2024 | Official leaderboard |
| 8 | | 55.9% | C_FDRS | Apr 2026 | Official leaderboard |
| 9 | | 54.3% | X_FD | Nov 2024 | Official leaderboard |
| 10 | | 52% | C_FDRS | Sep 2025 | Official leaderboard |
| 11 | Qwen2.5-Coder-14B-Instruct | 51.1% | X_FDRS | Nov 2024 | Official leaderboard |
| 12 | | 36.5% | X_FDRS | Jul 2024 | Official leaderboard |
| 13 | Qwen2.5-Coder-7B-Instruct | 32.8% | C_FDRS | Official leaderboard | |
| 14 | Qwen2.5-Coder-1.5B-Instruct | 9.90% | C_FX | Official leaderboard | |
| 15 | Qwen2.5-Coder-0.5B-Instruct | 0.60% | X_FD | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 68.7% | C_FDRS | Jul 2024 | Official leaderboard |
| 2 | | 67.3% | X_FDRS | Jul 2025 | Official leaderboard |
| 3 | Claude Haiku 4.5 | 63.8% | C_FDRS | Oct 2025 | Official leaderboard |
| 4 | GPT-5.4 Mini | 62.1% | C_FDRS | Mar 2026 | Official leaderboard |
| 5 | | 60.9% | X_FDRS | Feb 2026 | Official leaderboard |
| 6 | gpt-oss-120b | 60.3% | C_FDRS | Aug 2025 | Official leaderboard |
| 7 | | 57.5% | C_FDRS | Jul 2024 | Official leaderboard |
| 8 | | 55.9% | C_FDRS | Apr 2026 | Official leaderboard |
| 9 | | 54.3% | X_FD | Nov 2024 | Official leaderboard |
| 10 | | 52% | C_FDRS | Sep 2025 | Official leaderboard |
| 11 | Qwen2.5-Coder-14B-Instruct | 51.1% | X_FDRS | Nov 2024 | Official leaderboard |
| 12 | | 36.5% | X_FDRS | Jul 2024 | Official leaderboard |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official PowerCodeBench leaderboard
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official PowerCodeBench leaderboard
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official PowerCodeBench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official PowerCodeBench 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 Coding
- SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
- LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests.
DeepSeek-V4-Pro 93.5%
- SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
- DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
- SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases.
Seed2.1 Pro 59.8%
- Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance.
DeepSeek V4.1 Flash 3,471
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash. 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 official leaderboard (15 entries).
- Pass Models scored 12 models we track, plus 3 other entries on its leaderboard.
- Warning Newest score The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.4 Mini (#74).
- Pass Headroom The best is 68.7%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 148 rows for models we track and 3 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.