live
weekly refresh
basedagi.org
benchmark evidence

BenchLM Coding

BenchLM composite coding score (0-100).

winner on BenchLM Coding
direct benchmark result, not a broad vertical composite | source row dated 2026-05-28
scored on 2026-05-28
latest mapped results | top 20
#ModelScoreEvidenceTested
1OpenAI: GPT-5.4
Openai
87.8
model-only
aggregator
2026-05-28
2Anthropic: Claude Opus 4.6
Anthropic
85.9
model-only
aggregator
2026-05-28
3OpenAI: GPT-5.5
Openai
84.3
model-only
aggregator
2026-05-28
4DeepSeek: DeepSeek V4 Flash
Deepseek
82.7
model-only
aggregator
2026-05-28
5Anthropic: Claude Sonnet 4.6
Anthropic
82.2
model-only
aggregator
2026-05-28
6OpenAI: GPT-5.2
Openai
80.1
model-only
aggregator
2026-05-28
7Anthropic: Claude Sonnet 4.5
Anthropic
78.7
model-only
aggregator
2026-05-28
8OpenAI: GPT-5.1
Openai
77.1
model-only
aggregator
2026-05-28
9Z.ai: GLM 5
Z Ai
76.5
model-only
aggregator
2026-05-28
10Anthropic: Claude Opus 4.5
Anthropic
76.4
model-only
aggregator
2026-05-28
11DeepSeek: DeepSeek V4 Pro
Deepseek
69.7
model-only
aggregator
2026-05-28
12Google: Gemini 2.5 Pro
Google
49.3
model-only
aggregator
2026-05-28
what this result means

BenchLM composite coding score (0-100).

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

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

source record
category: coding
metric: score
matched models: 12
latest source date: 2026-05-28
direction: higher is better
inspect upstream source ->