Skip to content
Alpha
  • Models
  • Rankings
  • Ori

Footer

OpenRouterOpenRouter
© 2026 OpenRouter, Inc

Product

  • Chat
  • Rankings
  • Benchmarks
  • Apps
  • Discover
  • Models
  • Ori
  • Collections
  • Providers
  • Tools
  • Pricing
  • Business
  • EU routing
  • Enterprise
  • Labs

Company

  • About
  • Blog
  • Careers
    Hiring
  • Privacy
  • Terms of Service
  • Trust Center
  • Support
  • Works With OR
  • Data
  • Brand

Developer

  • Documentation
  • API Reference
  • Developer Platform
  • Status
  • AI Site Map

Connect

  • Discord
  • GitHub
  • LinkedIn
  • X
  • YouTube
Favicon for google

Google: Gemma 4 31B

google/gemma-4-31b-it

Model weights
Compare

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.

Modalities
In / Out Price
$0.10 / $0.34per 1M
Context
262K
Released
Apr 2, 2026
Compare
ProvidersPricingPerformanceUptimeAppsActivityFAQExplore

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Not currently available through EU in-region routing. In-region routing docs

Explore more models

AI Models with Vision: Multimodal LLMs for Image UnderstandingCollectionAI Model RankingsRanking
Latency / throughput
$0.100$0.340$0.10000.87s46 tps
100.00%
Turbo
$0.135$0.510$0.07500.68s39 tps
100.00%
$0.750$1.000$0.20000.22s167 tps
100.00%
Priority
$0.300$0.600--0.93s27 tps
98.96%
$0.140$0.400--0.78s105 tps
100.00%
$0.120$0.360$0.09000.67s38 tps
100.00%
$0.150$0.400$0.06000.83s25 tps
99.73%
$0.200$0.400--0.77s30 tps
99.74%
5% off
$0.38$0.361$1.15$1.093$0.19$0.18050.73s35 tps
98.20%
$0.120$0.370$0.01203.64s20 tps
96.01%
$0.380$1.150--1.52s82 tps
97.69%
$0.750$1.000$0.25001.71s9 tps
99.45%
$0.140$0.400$0.14000.41s52 tps
99.65%
$0.140$0.400--0.69s57 tps
90.67%
Not used in Standard routing:Why these endpoints are not used
Flex
$0.160$0.320--2.38s31 tps
99.79%
Throughput
167tok/s

P50, best across providers

Latency
0.22s

P50, best provider

AutoExacto benchmarks
GPQA DiamondTAU-BenchDekaLLM86.2%--Crusoe85.9%--SiliconFlow86.0%79.5%DeepInfra (fp8)84.9%79.7%ModelRun [by Modular]85.1%75.5%DeepInfra (fp4)84.5%76.0%
+9 more providers
Uptime (3d)The model was reachable. Request routed to a provider.
100.00%
Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.
99.92%

Availability over the last 3 days

Last 72 hours
Availability 99.92%
3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability
99.93%
Without Routing
94.18%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

1.
Favicon for https://www.mydreamcompanion.com/
My Dream Companion
new
43.2Btokens
2.
Favicon for https://raylu.ai/
Raylu
new
26.7Btokens
3.
Favicon for https://fiveblocks.com/
AIQ prompt eval
new
24.8Btokens
4.
Favicon for https://github.com/SkyrimNet-GamePlugin
SkyrimNet
new
19.1Btokens
5.
Favicon for https://kime.ai/
Kime
new
18.4Btokens

Frequently asked questions

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.

Gemma 4 31B costs $0.10/M input tokens and $0.34/M output tokens, with separate rates for Cache Read at $0.10/M tokens.

Gemma 4 31B has a 262,144 token context window. It supports up to 262,144 completion tokens.

Yes. Gemma 4 31B accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Gemma 4 31B accepts images, text, and video as input and returns text.

Gemma 4 31B is served by 12 providers on OpenRouter: CoreWeave, Venice, Chutes, DeepInfra (fp8), Crusoe, Friendli, NovitaAI, Parasail and 4 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

Gemma 4 31B was released on April 2, 2026.