The PHP subset of MultiPL-E: HumanEval and MBPP function-completion problems translated into PHP and scored with pass@1.
unassessed
| Category | coding |
|---|---|
| Subcategory | multilingual code generation |
| Page status | active |
| Metric | pass@1 |
| Direction | higher_is_better |
| Unit | % |
| Dataset licence | MIT |
| Publisher | Northeastern University Programming Research Lab (nuprl) |
This subset translates the HumanEval and MBPP prompts into PHP by rewriting each problem's function signature, docstring and tests with PHP syntax and typing, then asks the model to complete the function body in PHP. The underlying algorithmic problem is unchanged from the Python original; only the surface language differs.
Function completion in PHP: given a translated signature, docstring and (for HumanEval-derived items) doctests, the model generates a function body, which is compiled or interpreted with a real PHP toolchain inside a container and checked against translated unit tests.
| Model | Provider | Score | Card as of |
|---|---|---|---|
| GPT-5.1 | OpenAI | 86.5 | 2026-04 |
| GPT-5.1 Chat | OpenAI | 86.5 | 2026-04 |
| GPT-5.1 Codex | OpenAI | 86.5 | 2026-04 |
| GPT-5.1 Codex Max | OpenAI | 86.5 | 2026-04 |
| GPT-5.1 Codex mini | OpenAI | 86.5 | 2026-04 |
| Claude Opus 4 | Anthropic | 85.5 | 2026-04 |
| Claude Opus 4.6 | Anthropic | 85.5 | 2026-04 |
| GPT-5 | OpenAI | 85.2 | 2026-04 |
| GPT-5 Chat (latest) | OpenAI | 85.2 | 2026-04 |
| GPT-5 Mini | OpenAI | 85.2 | 2026-04 |
| GPT-5 Nano | OpenAI | 85.2 | 2026-04 |
| GPT-5 Pro | OpenAI | 85.2 | 2026-04 |
| GPT-5-Codex | OpenAI | 85.2 | 2026-04 |
| GPT-5.2 | OpenAI | 85.2 | 2026-04 |
| GPT-5.2 Chat | OpenAI | 85.2 | 2026-04 |
| GPT-5.2 Codex | OpenAI | 85.2 | 2026-04 |
| GPT-5.2 Pro | OpenAI | 85.2 | 2026-04 |
| GPT-5.3 Chat (latest) | OpenAI | 85.2 | 2026-04 |
| GPT-5.3 Codex | OpenAI | 85.2 | 2026-04 |
| GPT-5.3 Codex Spark | OpenAI | 85.2 | 2026-04 |
| GPT-5.4 | OpenAI | 85.2 | 2026-04 |
| GPT-5.4 mini | OpenAI | 85.2 | 2026-04 |
| GPT-5.4 nano | OpenAI | 85.2 | 2026-04 |
| GPT-5.4 Pro | OpenAI | 85.2 | 2026-04 |
| o3 | OpenAI | 85.1 | 2026-04 |
| Claude Opus 4.5 | Anthropic | 84.8 | 2026-04 |
| Claude Opus 4.5 (latest) | Anthropic | 84.8 | 2026-04 |
| Qwen3-Coder 480B-A35B Instruct | Alibaba / Qwen Team | 84.2 | 2026-04 |
| Claude Opus 4.1 | Anthropic | 83.5 | 2026-04 |
| Claude Opus 4.1 (latest) | Anthropic | 83.5 | 2026-04 |
| Claude Sonnet 4.6 | Anthropic | 83.5 | 2026-04 |
| GPT-4.1 | OpenAI | 82.8 | 2026-04 |
| Grok 4 | xAI | 82.8 | 2026-04 |
| Grok 4 Fast | xAI | 82.8 | 2026-04 |
| Grok 4 Fast (Non-Reasoning) | xAI | 82.8 | 2026-04 |
| Grok 4.1 Fast | xAI | 82.8 | 2026-04 |
| Grok 4.1 Fast (Non-Reasoning) | xAI | 82.8 | 2026-04 |
| Grok 4.20 (Non-Reasoning) | xAI | 82.8 | 2026-04 |
| Grok 4.20 (Reasoning) | xAI | 82.8 | 2026-04 |
| Grok 4.20 Multi-Agent | xAI | 82.8 | 2026-04 |
| Gemini 2.5 Pro | Google DeepMind | 82.5 | 2026-04 |
| Claude Sonnet 4 | Anthropic | 82.2 | 2026-04 |
| Claude Sonnet 4.5 | Anthropic | 82.2 | 2026-04 |
| Claude Sonnet 4.5 (latest) | Anthropic | 82.2 | 2026-04 |
| o4-mini | OpenAI | 82.1 | 2026-04 |
| o4-mini-deep-research | OpenAI | 82.1 | 2026-04 |
| Claude Opus 4 (latest) | Anthropic | 81.8 | 2026-04 |
| Qwen3 235B-A22B | Alibaba / Qwen Team | 81.2 | 2026-04 |
| DeepSeek V3.2 Exp | DeepSeek | 80.8 | 2026-04 |
| GPT-4o | OpenAI | 80.5 | 2026-04 |
| GPT-4o (2024-05-13) | OpenAI | 80.5 | 2026-04 |
| GPT-4o (2024-08-06) | OpenAI | 80.5 | 2026-04 |
| GPT-4o (2024-11-20) | OpenAI | 80.5 | 2026-04 |
| DeepSeek V3.2 | DeepSeek | 80.2 | 2026-04 |
| o3-mini | OpenAI | 80.2 | 2026-04 |
| Claude Sonnet 4 (latest) | Anthropic | 79.8 | 2026-04 |
| DeepSeek R1 0528 | DeepSeek | 79.5 | 2026-04 |
| DeepSeek R1 0528 NVFP4 v2 | NVIDIA | 79.5 | 2026-04 |
| Claude Sonnet 3.7 | Anthropic | 78.8 | 2026-04 |
| GPT-4 Turbo | OpenAI | 78.2 | 2026-04 |
| DeepSeek R1 | DeepSeek | 77.8 | 2026-04 |
| DeepSeek Reasoner | DeepSeek | 77.8 | 2026-04 |
| Claude Sonnet 3.5 | Anthropic | 77.5 | 2026-04 |
| Claude Sonnet 3.5 v2 | Anthropic | 77.5 | 2026-04 |
| DeepSeek V3.1 | DeepSeek | 77.2 | 2026-04 |
| Gemini 2.5 Flash | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Image | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Image (Preview) | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Lite | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Lite Preview 06-17 | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Lite Preview 09-25 | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Preview 04-17 | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Preview 05-20 | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Preview 09-25 | Google DeepMind | 76.2 | 2026-04 |
| Gemini 2.5 Flash Preview TTS | Google DeepMind | 76.2 | 2026-04 |
| Grok 3 | xAI | 76.2 | 2026-04 |
| Grok 3 Fast | xAI | 76.2 | 2026-04 |
| Grok 3 Fast Latest | xAI | 76.2 | 2026-04 |
| Grok 3 Latest | xAI | 76.2 | 2026-04 |
| Grok 3 Mini | xAI | 76.2 | 2026-04 |
| Grok 3 Mini Fast | xAI | 76.2 | 2026-04 |
| Grok 3 Mini Fast Latest | xAI | 76.2 | 2026-04 |
| Grok 3 Mini Latest | xAI | 76.2 | 2026-04 |
| Claude Haiku 4.5 | Anthropic | 75.8 | 2026-04 |
| Claude Haiku 4.5 (latest) | Anthropic | 75.8 | 2026-04 |
| Qwen3 32B | Alibaba / Qwen Team | 75.8 | 2026-04 |
| Qwen3 32B AWQ | Alibaba / Qwen Team | 75.8 | 2026-04 |
| Qwen3 32B NVFP4 | NVIDIA | 75.8 | 2026-04 |
| Mistral Large (latest) | Mistral AI | 75.5 | 2026-04 |
| Mistral Large 2.1 | Mistral AI | 75.5 | 2026-04 |
| Mistral Large 3 | Mistral AI | 75.5 | 2026-04 |
| DeepSeek V3 0324 | DeepSeek | 74.8 | 2026-04 |
| GPT-4 | OpenAI | 74.2 | 2026-04 |
| GPT-4.1 mini | OpenAI | 74.2 | 2026-04 |
| GPT-4.1 nano | OpenAI | 74.2 | 2026-04 |
| Qwen2.5 Coder 32B Instruct | Alibaba / Qwen Team | 74.2 | 2026-04 |
| Qwen2.5 Coder 32B Instruct AWQ | Alibaba / Qwen Team | 74.2 | 2026-04 |
| DeepSeek Chat | DeepSeek | 73.5 | 2026-04 |
| DeepSeek V3 | DeepSeek | 73.5 | 2026-04 |
| Llama 4 Maverick 17B 128E Instruct | Meta | 73.2 | 2026-04 |
| Llama-4-Maverick-17B-128E-Instruct-FP8 | Meta | 73.2 | 2026-04 |
| Claude Opus 3 | Anthropic | 72.8 | 2026-04 |
| Qwen2.5 72B Instruct | Alibaba / Qwen Team | 72.2 | 2026-04 |
| Gemma 4 31B | Google DeepMind | 71.5 | 2026-04 |
| gemma 4 31B it | Google DeepMind | 71.5 | 2026-04 |
| gemma 4 31B it GGUF | Unsloth | 71.5 | 2026-04 |
| Gemma 4 31B IT NVFP4 | NVIDIA | 71.5 | 2026-04 |
| Codestral (latest) | Mistral AI | 71.2 | 2026-04 |
| Llama 4 Scout 17B 16E | Meta | 70.5 | 2026-04 |
| Llama 4 Scout 17B 16E Instruct | Meta | 70.5 | 2026-04 |
| Llama-4-Scout-17B-16E-Instruct-FP8 | Meta | 70.5 | 2026-04 |
| GPT-4o mini | OpenAI | 70.2 | 2026-04 |
| Llama 3.3 70B Instruct NVFP4 | NVIDIA | 70.2 | 2026-04 |
| Llama-3.3-70B-Instruct | Meta | 70.2 | 2026-04 |
| Mistral Medium (latest) | Mistral AI | 70.2 | 2026-04 |
| Mistral Medium 3 | Mistral AI | 70.2 | 2026-04 |
| Mistral Medium 3.1 | Mistral AI | 70.2 | 2026-04 |
| Gemma 4 26B | Google DeepMind | 69.8 | 2026-04 |
| phi 4 | Microsoft | 69.5 | 2026-04 |
| Claude Haiku 3.5 | Anthropic | 68.8 | 2026-04 |
| Claude Haiku 3.5 (latest) | Anthropic | 68.8 | 2026-04 |
| Llama 3.1 70B | Meta | 68.5 | 2026-04 |
| Llama 3.1 70B Instruct | Meta | 68.5 | 2026-04 |
| Qwen2.5 Coder 14B Instruct | Alibaba / Qwen Team | 66.5 | 2026-04 |
| Mistral Small (latest) | Mistral AI | 62.5 | 2026-04 |
| Mistral Small 24B Instruct 2501 | Mistral AI | 62.5 | 2026-04 |
| Mistral Small 3.1 24B Instruct 2503 | Mistral AI | 62.5 | 2026-04 |
| Mistral Small 3.2 | Mistral AI | 62.5 | 2026-04 |
| Mistral Small 3.2 24B Instruct 2506 | Mistral AI | 62.5 | 2026-04 |
| Mistral Small 4 | Mistral AI | 62.5 | 2026-04 |
| Mistral Small 4 119B 2603 | Mistral AI | 62.5 | 2026-04 |
| Qwen2.5 Coder 7B Instruct | Alibaba / Qwen Team | 59.0 | 2026-04 |
| Qwen2.5 Coder 7B Instruct GPTQ Int4 | Alibaba / Qwen Team | 59.0 | 2026-04 |
| Phi 4 mini instruct | Microsoft | 58.8 | 2026-04 |
| CodeLlama 34B Instruct hf | Meta | 52.8 | 2026-04 |