DEV Community

#jax

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Gemma 4 E2B in JAX on One TPU v5e: bf16 or 4-Bit, Python or Rust, and How It Stands Against vLLM
xbill profile

Gemma 4 E2B in JAX on One TPU v5e: bf16 or 4-Bit, Python or Rust, and How It Stands Against vLLM

#jax #tpu #gemma #rust
10
Comments
10 min read
Gemma 4 E2B in Pure JAX on a Colab TPU: Google's 4-Bit Export Against an Exact Repack

A repack beats Google's export by 342x
xbill profile

Gemma 4 E2B in Pure JAX on a Colab TPU: Google's 4-Bit Export Against an Exact Repack

#jax #tpu #gemma #machinelearning
2
Comments 4
8 min read
Installing JAX with ROCm Acceleration on Ubuntu 24.04
sharmukh profile

Installing JAX with ROCm Acceleration on Ubuntu 24.04

#jax #rocm #machinelearning #ubuntu
7
Comments
2 min read
Gemma 4 in Pure JAX: What Ports from TPU to GPU, and What Doesn't
xbill profile

Gemma 4 in Pure JAX: What Ports from TPU to GPU, and What Doesn't

#gemma #jax #googleai #machinelearning
8
Comments 3
6 min read
g5g vs g6 for LLM Serving: the Same Code, and 3.7x the Throughput
xbill profile

g5g vs g6 for LLM Serving: the Same Code, and 3.7x the Throughput

#aws #jax #cuda #machinelearning
5
Comments 2
6 min read
Gemma 4 in Pure JAX: What Changes Between Turing and Ada, and What Doesn't
xbill profile

Gemma 4 in Pure JAX: What Changes Between Turing and Ada, and What Doesn't

#jax #gemma #cuda #machinelearning
3
Comments
8 min read
Deploying a QAT Checkpoint Your Serving Stack Can't Load: Gemma 4 E2B in Pure JAX on One TPU
xbill profile

Deploying a QAT Checkpoint Your Serving Stack Can't Load: Gemma 4 E2B in Pure JAX on One TPU

#tpu #jax #llm #quantization
8
Comments
10 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.