BenchMark'd
No ratings yet
Be the first to review this model
Meta's large-scale multimodal model with 128 experts and 400B total parameters, designed for general assistant and chat use cases. Exceeds GPT-4o and Gemini 2.0 on coding, reasoning, multilingual, and image benchmarks. The most capable open-weight model available for demanding production workloads.
Released
April 5, 2025
Parameters
400B (MoE, 17B active)
Context
1M
Pricing
Open Source
Last updated: March 15, 2026
Benchmark scores may vary based on evaluation methodology and conditions.