live
weekly refresh
basedagi.org
benchmark evidence

EnigmaEval

Scale Labs puzzle-solving benchmark for long-form reasoning over public puzzle archives.

winner on EnigmaEval
direct benchmark result, not a broad vertical composite | source row dated 2025-08-08
scored on 2025-12-15 · stale source data (165d)
latest mapped results | top 20
#ModelScoreEvidenceTested
1OpenAI: GPT-5
Openai
38.7
model-only
independent_benchmark
2025-08-08
2OpenAI: GPT-5.2
Openai
38.4
model-only
independent_benchmark
2025-12-15
3Anthropic: Claude Opus 4.5
Anthropic
17.2
model-only
independent_benchmark
2025-11-26
4Anthropic: Claude Sonnet 4.5
Anthropic
12.5
model-only
independent_benchmark
2025-10-02
5Anthropic: Claude Opus 4.7
Anthropic
11.9
model-only
independent_benchmark
2025-05-23
6Anthropic: Claude Sonnet 4.6
Anthropic
8.1
model-only
independent_benchmark
2025-05-23
7OpenAI: GPT-4.1
Openai
8.0
model-only
independent_benchmark
2025-03-05
8Meta: Llama 4 Maverick
Meta Llama
2.1
model-only
independent_benchmark
2025-04-05
what this result means

Scale Labs puzzle-solving benchmark for long-form reasoning over public puzzle archives.

This benchmark contributes direct public evidence. Read its scope before generalizing the result.

A win here is a win on EnigmaEval. Broad task pages require independent corroboration before naming a general winner.

source record
category: reasoning
metric: accuracy
matched models: 8
latest source date: 2025-12-15
direction: higher is better
inspect upstream source ->