VECTOR WIREAI INTELLIGENCE
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models Deals Regulatory Sources

Claude 4 Opus vs Llama 3.1 Instruct 405B

AnthropicvsMeta29 shared benchmarks272 head-to-head
BenchmarkClaude 4 OpusLlama 3.1 Instruct 405B
AA Intelligence31.78.5
AA-LCR4024.7
aider_polyglot70.75.8
AIME 20247623.3
AIR-Bench 202485.758.6
anthropic_red_team98.996.5
Artificial Analysis Coding Index3414.5
bbq99.394.5
Fortress27.620.6
GPQA Diamond79.651.5
harmbench91.762.7
HLE12.34.2
IFBench53.739
ifeval87.488.6
LiveCodeBench70.430.1
longbench_v255.636.1
math_500_em98.273.8
mmlu92.988.6
mmlu_redux94.286.2
MMLU-Pro86.673.4
mmmlu88.873.8
scicode40.929.9
simple_safety_tests10098.8
simplebench58.823
simpleqa22.823.2
SWE-bench Verified79.424.5
Terminal-Bench Hard31.16.8
xstest9795.9
τ²-Bench Telecom (AA run)73.419

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.