Skip to content

Benchmark

t2-bench

Scores are percentages.

Models measured
21
Scores recorded
21
Sources
1
IQ weight
2.24

Score distribution

73.7%
Mean
80.2%
Median
20.6%
Std dev
48.8%
10th percentile
89.7%
90th percentile
11.6%
Min
11.6%99.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.24 for chat models, measured over 21 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 3.1 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim99.3%
2Gemini 3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim90.2%
3GLM 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.7%
4Qwen3.5 397B A17B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim86.7%
5Gemma 4 31B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim86.4%
6Gemma 4 26B A4B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim85.5%
7Gemini 3 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim85.4%
8Qwen3.5 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim81.2%
9DeepSeek V3.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.3%
10DeepSeek V3.2 Speciale· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.3%
11DeepSeek Reasoner· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.2%
12Qwen3.5 4B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79.9%
13Qwen3.5 122B-A10B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79.5%
14Qwen3.5 9B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79.1%
15Qwen3.5 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79%
16Qwen3 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.8%
17GPT OSS 120Bhigh· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim63.9%
18Gemma 4 E4B-it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim57.5%
19Qwen3.5 2B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim48.8%
20Gemma 4 E2B-it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim29.4%
21Qwen3.5 0.8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim11.6%

Data credits: Data by LLM Stats