Skip to content

Benchmark

Multi-Challenge

Instruction following benchmarks measure how closely models follow instructions. Scores are percentages. Higher is better.

Models measured
30
Scores recorded
33
Sources
2
IQ weight
1.22

Score distribution

52.6%
Mean
55.2%
Median
14.6%
Std dev
35.8%
10th percentile
67.6%
90th percentile
15%
Min
15%76.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.22 for chat models, measured over 30 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Nova 2 Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim76.6%
2Nova 2 LiteAmazon Nova 2 Lite official launch chart / technical report · accuracy, v1· Reported by Amazon Web Services, Amazon AGI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.76.6%
3GPT 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.6%
4Qwen3.5 397B A17B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67.6%
5Qwen3.5 397B A17BQwen3.5 release/model card · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.6%
6Doubao Seed 1.8· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim66.7%
7Nemotron 3 Ultra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim63.8%
8Qwen3 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim63.3%
9Qwen3.5 122B-A10B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim61.5%
10Seed 2.0 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim61.1%
11Qwen3.5 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60.8%
12Step 3.5 FlashStep 3.5 technical report · Table 5 · PaCoRe· Reported by StepFun· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.8%
13OpenAI o3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60.4%
14Mercury 3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60.4%
15Qwen3.5 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60%
16Step 3.5 FlashStep 3.5 technical report · Table 5 · vanilla· Reported by StepFun· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.55.7%
17Nemotron 3 Super 120B A12B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim55.2%
18Qwen3.5 9B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim54.5%
19Kimi K2 Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim54.1%
20Mai Thinking 1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim53%
21Qwen3.5 4B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim49%
22MiniMax M1 40K· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.44.7%
23MiniMax M1 80K· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.44.7%
24GPT-4.5 Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim43.8%
25o4 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim43%
26GPT 4o· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim40.3%
27o3 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim39.9%
28Nemotron 3 Nano 30B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim38.5%
29GPT 4.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim38.3%
30GPT-4.1 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim35.8%
31Qwen3.5 2B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim33.7%
32Qwen3.5 0.8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim18.9%
33GPT-4.1 nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim15%

Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.