---
id: "google/gemma-4-31b-it"
also: ["google/gemma-4-31b-it-20260402", "google/gemma-4-31B-it", "hf/google/gemma-4-31B-it", "requesty/google/gemma-4-31b-it", "vercel/google/gemma-4-31b-it"]
cheapest_list_usd_per_m: "0.08 in, 0.35 out, at open-inference/bf16 on openrouter"
cheapest_effective_usd_per_m: "0.0798 in, 0.3485 out, at open-inference/bf16 on openrouter"
context: 262144
max_output: 16384
input: ["image", "text", "video"]
output: ["text"]
released: "2026-04-02"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logit_bias", "logprobs", "max_completion_tokens", "max_tokens", "min_p", "presence_penalty", "reasoning", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 44
gateways: "openrouter (29), vercel (8), hf (5), requesty (2)"
no_training: 30
trains: 3
retains_prompts: 7
intelligence: 4.4000
coding: 43.4000
agentic: 4.2000
---
# Google: Gemma 4 31B

Gemma 4 31B is engineered to tackle the most demanding enterprise workloads and complex reasoning tasks. With an expansive 256K-token context window, the 31B model can effortlessly ingest entire codebases, and massive sets of images in a single prompt.