BREAKING
Gemma 4 31B Hits 1,851 Tokens/sec
0
tok/s
Throughput
0
x
vs typical GPU
0
s
Time to first token
Open Speech-to-Speech Stack
1
Parakeet ASR
↓
2
Gemma 4 on Cerebras
↓
3
Text-to-speech
Gemma 4 31B Benchmarks
MMLU Pro
85.2
AIME 2026
89.2
LiveCodeBench
80
MMMU Pro
76.9
Strengths and Caveats
Upsides
Praise
●
Fluid voice demo
●
Stable JSON output
●
Apache 2.0 license
Caveats
Flagged
●
Reasoning stalls on hard prompts
●
Heavy local execution
●
Quantized GGUF builds help
Open Voice Agents Go Real-Time
AI NEWS BLITZ
Google's open Gemma 4 31B now powers near-instant voice AI on Cerebras hardware.