Skip to content

Benchmark

BenchCAD

Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.

Models measured
14
Scores recorded
23
Sources
2
IQ weight
2.04

Score distribution

67.2%
Mean
70.6%
Median
15.6%
Std dev
44.4%
10th percentile
83.4%
90th percentile
27.3%
Min
27.3%95.9%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.04 for chat models, measured over 13 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 AstraOpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.95.9%
2Claude Fable 5.1OpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.84.3%
3GPT-5.6 SolPython tool enabled· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.83.4%
4GPT-5.6 SolOpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.83.3%
5Claude Opus 5OpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.82.1%
6GPT-5.6 TerraPython tool enabled· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.78.2%
7Claude Sonnet 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim74.7%
8GPT-5.6 LunaPython tool enabled· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.73.9%
9Claude Opus 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73%
10Kimi K3.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim72.8%
11GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim70.6%
12GPT-5.6 Solno Python tool· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.70.6%
13Claude Fable 5OpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.5%
14Claude Haiku 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67%
15GPT-5.6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim63.1%
16GPT-5.6 Lunano Python tool· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.63.1%
17GPT-5.6 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim62.3%
18GPT-5.6 Terrano Python tool· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.3%
19GPT 5.5Python tool enabled· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.55.8%
20Claude Opus 4.8Python tool enabled· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.51.8%
21GPT 5.5no Python tool· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.44.4%
22Claude Sonnet 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim37.3%
23Claude Opus 4.8no Python tool· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.27.3%

Data credits: 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. · Data by LLM Stats