lucataco/gemma-4-31b-it 🖼️🔢📝 → 📝

▶️ 282 runs 📅 May 2026 ⚙️ Cog 0.19.1 🔗 GitHub 📄 Paper ⚖️ License
image-to-text question-answering text-generation visual-understanding

Performance

18.8sTypical run time
~121sCold start (first call)
282Total runs

About

Gemma 4 31B Instruct - Google open-weight VLM (image + text in, text out)

Example Output

Prompt:

"Explain quantum computing in simple terms"

Output

To understand quantum computing, you first have to understand how a classical computer (the phone or laptop you are using right now) works.

1. The Classical Way: Bits

Classical computers use Bits. A bit is like a light switch: it can be either Off (0) or On (1).

Everything you do on a computer—typing an email, watching a video, playing a game—is just billions of these 0s and 1s flipping on and off in different combinations. It’s very fast, but it can only do one thing at a time: it processes a sequence of steps.

2. The Quantum Way: Qubits

Quantum computers use Qubits. These are based on the laws of quantum physics, and they have two "superpowers" that make them different from regular bits:

Superpower #1: Superposition (The Spinning Coin)

Imagine you flip a coin. While it’s lying on the table, it is either Heads (1) or Tails (0). That is a classical bit.

But imagine a coin spinning on the table. While it is spinning, it is sort of both heads and tails at

Performance Metrics

18.75s Prediction Time
120.95s Total Time
All Input Parameters
{
  "top_p": 0.95,
  "prompt": "Explain quantum computing in simple terms",
  "temperature": 0.7,
  "system_prompt": "",
  "max_new_tokens": 256
}
Input Parameters
image Type: stringDefault: null
Optional input image for multimodal queries.
top_p Type: numberDefault: 0.95Range: 0 - 1
Nucleus sampling top-p.
prompt Type: stringDefault: Explain quantum computing in simple terms.
User prompt. The chat template is applied automatically.
temperature Type: numberDefault: 0.7Range: 0 - 2
Sampling temperature. 0 = greedy decoding.
system_prompt Type: stringDefault:
Optional system prompt.
max_new_tokens Type: integerDefault: 256Range: 1 - 2048
Maximum number of tokens to generate.
Output Schema

Output

Type: string

Version Details
Version ID
14c757b42dc096c7d2926c72b40503f5d4acbf1b9b27e910e6722b2a96e51cde
Version Created
May 3, 2026
Run on Replicate →