Skip to main content

Gemma 4 31B

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.

Model card on Hugging Face
Input / output price
$0.00 / $0.00 per 1M
Context
262K tokens
Released
Apr 2, 2026
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "google_gemini_ai/gemma-4-31b-it".

Read docs

Providers

12 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
Google Gemini AIgoogle_gemini_ai/gemma-4-31b-it
$0.00/M$0.00/M
Featherlessfeatherless/gemma-4-31b-it
$0.12/M$0.36/M
Chuteschutes/gemma-4-31b-it
$0.12/M$0.37/M
Nextbit256nextbit256/gemma4:31b
$0.125/M$0.34/M
Concentrateconcentrate/gemma-4-31b
$0.14/M$0.40/M

Is a provider down? OpenRouter status

Pricing

Averaged across 12 providers, in USD per 1M tokens after gateway discounts.

Average input
$0.16 per 1M
Average output
$0.37 per 1M
Average cache read
$0.13 per 1M

Cost by volume

Total cost from 0 to 1T tokens on the cheapest and most expensive provider.

Highest$0.349/MIO.netLowest$0.00/MGoogle Gemini AI

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

112tokens/sacross providers

Faster than 81% of the other 599 models on the gateway

Time to first token

Lower is better

3.6sacross providers

Faster than 26% of the other 152 models on the gateway

Response latency

Lower is better

11sacross providers

Faster than 17% of the other 780 models on the gateway

Benchmarks

The Respan Index plus 15 published scores from 5 sources, each linked to where it was published.

18.7/ 100

Respan Index#140 of 176

176 ranked models. The shaded band is this model's 95% interval, 14.9 to 26.6. Qwen3 235B-A22B Thinking 2507 and Gemini 2.5 Flash fall inside it, so the gap is within the margin of error.
Gemma 4 31B by category: Coding 7.8, Agents 12.5 (estimated), Reasoning 19.7, Knowledge & language 11.0, Human preference 50.4
  1. 140Gemma 4 31B18.7
  2. 141Qwen3 235B-A22B Thinking 250718.5
  3. 141Gemini 2.5 Flash18.5

Published scores

BenchmarkScore
LiveCodeBench v6Google80%
WeirdMLHåvard Tveit Ihle52.3%

Compare Gemma 4 31B: Gemma 4 31B vs Gemini 4 Argon, Gemma 4 31B vs Gemma 4 26B A4B, Gemma 4 31B vs Gemini 4 Neon, Gemma 4 31B vs Gemini 4 Carbon, Gemma 4 31B vs Gemini 4 Hydrogen, Gemma 4 31B vs Gemini 3.8 Flash

Apps

Use Gemma 4 31B in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Gemma 4 31B FAQ

Short answers, built from the data on this page.

Gemma 4 31B is a model from Google Gemini AI, released on Apr 2, 2026. 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.

On the Respan gateway, Gemma 4 31B costs from $0.00 per 1M input tokens and $0.00 per 1M output tokens on Google Gemini AI, after gateway discounts. Averaged across its 12 providers, it costs $0.16 per 1M input tokens and $0.37 per 1M output tokens.

Gemma 4 31B is available through 12 providers: Google Gemini AI, Featherless, Chutes, Nextbit256, Concentrate, FriendliAI, and 6 more.

Gemma 4 31B accepts up to 262K tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "google_gemini_ai/gemma-4-31b-it". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Gemma 4 31B scores 18.7 out of 100 on the Respan Index v1.1, ranking #140 of 176 models. By category: Coding 7.8, Agents 12.5 (estimated), Reasoning 19.7, Knowledge & language 11.0, and Human preference 50.4. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

Gemma 4 31B has 15 published benchmark scores, including LiveCodeBench v6 80%, GPQA Diamond 84.3%, and AIME 2026 (no tools) 89.2%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.