benchmark evidence
UGI — ugi-natural-intelligence
UGI Leaderboard — ugi-natural-intelligence score from DontPlanToEnd/UGI-Leaderboard
activeupstream source ->
winner on UGI — ugi-natural-intelligence
OpenAI: GPT-5.579.3
direct benchmark result, not a broad vertical composite | source row dated 2026-04-25
scored on 2026-04-25 · stale source data (35d)
latest mapped results | top 20
| # | Model | Score | Evidence | Tested |
|---|---|---|---|---|
| 1 | OpenAI: GPT-5.5 | 79.3 | model-only independent_benchmark | 2026-04-25 |
| 2 | OpenAI: GPT-5.4 | 73.5 | model-only independent_benchmark | 2026-03-05 |
| 3 | OpenAI: GPT-5 | 69.3 | model-only independent_benchmark | 2025-09-24 |
| 4 | OpenAI: o3 | 66.2 | model-only independent_benchmark | 2025-09-24 |
| 5 | OpenAI: o1 | 64.2 | model-only independent_benchmark | 2025-09-24 |
| 6 | Google: Gemini 2.5 Pro | 64.0 | model-only independent_benchmark | 2025-09-09 |
| 7 | OpenAI: GPT-5.1 | 62.4 | model-only independent_benchmark | 2025-11-16 |
| 8 | OpenAI: GPT-5.2 | 61.5 | model-only independent_benchmark | 2026-03-04 |
| 9 | OpenAI: GPT-4.1 | 59.9 | model-only independent_benchmark | 2025-09-24 |
| 10 | MoonshotAI: Kimi K2 Thinking | 54.4 | model-only independent_benchmark | 2025-11-07 |
| 11 | DeepSeek: R1 | 52.0 | model-only independent_benchmark | 2025-09-21 |
| 12 | Anthropic: Claude Sonnet 4 | 51.1 | model-only independent_benchmark | 2025-09-12 |
| 13 | OpenAI: o4 Mini | 45.5 | model-only independent_benchmark | 2025-10-04 |
| 14 | Mistral Large | 37.6 | model-only independent_benchmark | 2025-09-30 |
| 15 | Qwen: Qwen3 235B A22B | 36.5 | model-only independent_benchmark | 2025-09-21 |
| 16 | Google: Gemma 3 27B | 34.5 | model-only independent_benchmark | 2025-09-06 |
| 17 | Meta: Llama 3.3 70B Instruct | 31.4 | model-only independent_benchmark | 2025-09-10 |
| 18 | Mistral: Mistral Nemo | 20.8 | model-only independent_benchmark | 2025-09-06 |
| 19 | Qwen: Qwen3 32B | 20.1 | model-only independent_benchmark | 2025-09-13 |
| 20 | Qwen: Qwen3 30B A3B | 19.2 | model-only independent_benchmark | 2025-11-08 |
what this result means
UGI Leaderboard — ugi-natural-intelligence score from DontPlanToEnd/UGI-Leaderboard
This benchmark contributes direct public evidence. Read its scope before generalizing the result.
A win here is a win on UGI — ugi-natural-intelligence. Broad task pages require independent corroboration before naming a general winner.
source record
category: uncensored
metric: accuracy
matched models: 20
latest source date: 2026-04-25
direction: higher is better