Skip to content

Benchmark

FrontierCode 1.1 Main

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

Models and systems measured
15
Scores recorded
57
Sources
1
IQ weight
No weight published

Score distribution

46%
Mean
47.7%
Median
7.8%
Std dev
37.1%
10th percentile
53.4%
90th percentile
15.4%
Min
15.4%54.6%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#Model or systemScore
1Claude Opus 5.5Opus 5.5 System Card capability section · Claude Code · medium effort · mean@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.54.6%
2Claude Opus 5.5Opus 5.5 System Card capability section · Claude Code · max effort · mean@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.54.4%
3Claude Fable 5OpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.5%
4Claude Fable 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.5%
5Claude Fable 5Opus 5.5 System Card capability section · best reported effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.5%
6Claude Opus 5medium effort · $4.31 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.4%
7Claude Opus 5OpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.4%
8Claude Opus 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.4%
9Claude Opus 5Opus 5.5 System Card capability section · best reported effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.4%
10GPT-6 Astrahighest effort· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.3%
11GPT-6 Astramax effort · $4.59 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.3%
12GPT-6 AstraOpus 5.5 System Card capability section · best reported effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.53.3%
13Claude Fable 5.1Opus 5.5 System Card capability section · best reported effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.52.8%
14Claude Sonnet 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · Claude Code · xhigh · mean@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.52.1%
15Claude Fable 5.1medium effort · $3.28 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.50.9%
16Claude 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.50.9%
17GPT-6 Astrahigh effort · $3.01 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.50.9%
18GPT-6 Astraxhigh effort · $3.28 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.50.6%
19Claude Fable 5.1high effort · $5.27 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.50.3%
20Claude Fable 5.1max effort · $12.83 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.50.3%
21Claude Fable 5.1low effort · $2.38 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.49.8%
22GPT-6 Solmax effort · $2.137 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.49.3%
23GPT-6 Astramedium effort · $2.43 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.48.8%
24Claude Fable 5.1xhigh effort · $9.27 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.48.7%
25GPT-6 Solxhigh effort · $1.3748 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.48.5%
26Claude Opus 5high effort · $7.24 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.48%
27Claude Opus 5max effort · $11.42 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.48%
28GPT-6 Solhigh effort · $1.0793 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.47.7%
29GPT-5.6 Solmax effort · $5.19 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.47.5%
30GPT-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.47.5%
31GPT-5.6 SolOpus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.47.5%
32GPT-5.6 Solxhigh effort · $4.15 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.46.8%
33Claude Opus 4.8Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.46.5%
34Claude Haiku 5.5Claude Code · max · mean@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.46.4%
35Claude Sonnet 5.5Claude Code · max · mean@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.46.2%
36GPT-6 Solmedium effort · $0.7966 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45.9%
37Claude Haiku 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · Claude Code · xhigh · mean@5· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45.8%
38GPT-6 Astralow effort · $1.7 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45.3%
39GPT-5.6 Solhigh effort · $3.48 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45.1%
40Claude Opus 5xhigh effort · $9.14 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.43.6%
41Gemini 3.7 FlashScore· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.43.6%
42Gemini 3.8 FlashOpenAI launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.43.6%
43GPT-6 Lunamax effort · $0.1072 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.42.4%
44Claude Opus 5low effort · $2.68 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.41.9%
45GPT-5.6 Solmedium effort · $2.69 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.39.9%
46GPT-5.6 Lunamax effort · $0.37 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.39.8%
47GPT-5.6 Lunaxhigh effort · $0.31 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.38.9%
48GPT-6 Sollow effort · $0.453 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.37.3%
49GPT-6 Lunahigh effort · $0.0672 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.37.3%
50GPT-6 Lunaxhigh effort · $0.0728 average cost/task· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.37.1%

Showing 1–50 of 56 results.

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
GPT-6 Sol (Sonnet 5.5 table label)source table label · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.49.3%

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.