Groq 3 LPX

product→ stable

Groq 3 LPX accelerator

NVIDIA’s Groq 3 LPX is a dedicated AI inference accelerator now in full production, designed for latency-sensitive agentic workloads. It delivers a world-class speed of up to 3,400 output tokens per second, extending the Vera Rubin NVL72 platform for

1Total Mentions

+0.60Sentiment (Very Positive)

+1.2%Velocity (7d)

View subgraph

First seen: Aug 24, 2026Last active: 6h ago

Signal Radar

Five-axis snapshot of this entity's footprint

live

Loading radar…

Mentions × Lab Attention

Weekly mentions (solid) and average article relevance (dotted)

mentionsrelevance

Loading timeline…

Timeline

Product LaunchAug 24, 2026
Groq 3 LPX hits 3,400 tokens/s on Gemma 4 31B, now in full production on Vera Rubin platform
View source
tokens per second:
3,400

Relationships

Developed

←
Nvidia
company1 source95% conf.

Uses

←
Nebius
company1 source70% conf.
←
Gemma 4 2B
ai model1 source50% conf.

Predictions

No predictions linked to this entity.

AI Discoveries

No AI agent discoveries for this entity.

Sentiment History

Positive sentiment

Negative sentiment

Range: -1 to +1