Skip to content

Benchmark

Agents' Last Exam

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models measured
32
Scores recorded
32
Sources
2
IQ weight
0.00

Score distribution

38.8%
Mean
35.4%
Median
14.8%
Std dev
25.2%
10th percentile
56.4%
90th percentile
12.4%
Min
12.4%78.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.00 for chat models, measured over 32 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT-6.2 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.3%
2GPT-6.2 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73%
3GPT-6.1 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60%
4GPT-6 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim59.3%
5GPT-6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim56.4%
6GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim52.7%
7Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim52.4%
8Qwen 3.8 Flash Next· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
51.2%
9Qwen3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim51.2%
10GPT-6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim50.9%
11GPT-5.6 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim50.4%
12GPT-5.6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim50.3%
13Kimi K3.1 Pro· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
43.2%
14Qwen 3.8 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
42.9%
15Doubao Seed 2.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim41.4%
16Gemini 4 Argon· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim39.5%
17DeepSeek V4.1 Flash 1017· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
35.4%
18DeepSeek V4.1 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim31.8%
19MiMo V2.6 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim31.6%
20MiMo V2.6 Pro RL· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim31.6%
21MiniMax M3.1 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
29.6%
22ERNIE 5.5 Flash Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim29%
23GLM 5.3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim28.5%
24Kimi K3· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim28.3%
25MiMo V2.6 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
27.6%
26DeepSeek V4 Flash Vision Exp· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim27.3%
27Gemini 3.7 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim26.3%
28GLM 5.3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim26.3%
29DeepSeek V4 Pro 0813· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim25.7%
30DeepSeek V4 Flash 0731· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim25.2%
31Hy4 preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim22.8%
32DeepSeek V4 Pro· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim12.4%

Data credits: Data by LLM Stats · Hugging Face official benchmark leaderboards; licence varies by benchmark