← Back to Directory

Gemma 4 31B

GOOGLE Developer Architecture Profile

OpenRouter ↗HuggingFace ↗
Intelligence (ELO)1583Chatbot Arena Verified
Max Context262,144Tokens
API Cost / 1M$0.54Blended Prompt + Completion

Model Capabilities

  • Top Tier
  • Coding & Logic
  • Fictional
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function calling, and multilingual support across 140+ languages. Strong on coding, reasoning, and document understanding tasks. Apache 2.0 license.

Granular Pricing Matrix

Input Tokens (Prompt)$0.14 / 1M
Output Tokens (Completion)$0.40 / 1M

Pricing data via OpenRouter. Sync: 4/2/2026

Evaluate Competitors

VS Engine MatchupGemma 4 31B vs xAI: Grok 4.20VS Engine MatchupGemma 4 31B vs DeepSeek: DeepSeek V3.2VS Engine MatchupGemma 4 31B vs Google: Lyria 3 Pro PreviewVS Engine MatchupGemma 4 31B vs xAI: Grok 4.20 Multi-AgentVS Engine MatchupGemma 4 31B vs DeepSeek: DeepSeek V3.2 SpecialeVS Engine MatchupGemma 4 31B vs Anthropic: Claude Sonnet 4.6