VECTOR WIREAI INTELLIGENCE
PKT
Refresh Models Deals Regulatory Sources

Questions this page answers

Which is better, Claude 3.5 Sonnet or InternVL-2-8B?
Across 10 shared benchmarks, Claude 3.5 Sonnet scores higher on 6 and InternVL-2-8B on 4. The widest gap is MMMU (val) (Pass@1), where Claude 3.5 Sonnet scores 68.3 against 46.3.

Claude 3.5 Sonnet vs InternVL-2-8B

Across 10 shared benchmarks, Claude 3.5 Sonnet scores higher on 6 and InternVL-2-8B on 4. The widest gap is MMMU (val) (Pass@1), where Claude 3.5 Sonnet scores 68.3 against 46.3.

AnthropicvsOpenGVLab10 shared benchmarks64 head-to-head
BenchmarkClaude 3.5 SonnetInternVL-2-8B
AI2D94.781.4
BLINK56.545.4
InterGPS (test)45.653.2
MathVista67.751.1
MMBench (dev-en)82.387
MMMU (val) (Pass@1)68.346.3
POPE (test)76.684.2
ScienceQA (img-test)73.895.9
TextVQA (val)70.568.8
Video-MME Overall55.952.6

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.