Roboflow

GoogleGoogleReleased Apr 2026Apache 2.0

Gemma 4 31B

Gemma 4 31B scores 57.7% overall (4 of 5 tasks) on Roboflow's Vision Evals (November 19, 2026 snapshot), #49 of 73 models (Gemini 4 Neon leads at 91.0%).

Its weakest task is identification (79.2%, #62 of 73). It is an open-weight model from Google under the Apache 2.0 license, and free for commercial use.

Vision Evals (4 of 5 tasks)57.7%Field median 65.7% · #49 of 73
Cost$0.0007/ sampleField median $0.0040 · #22 of 73
Speed34.4s/ sampleField median 10.2s · #68 of 73

Gemma 4 31B Interactive Demo

Model settings

Thinking level

Max output tokens

Default 16,384 · max 16,384

Sign in to adjust thinking and output length per run.

Results appear here. Add an image or pick an example to run Gemma 4 31B.

Gemma 4 31B Vision Evals

Vision Evals is Roboflow's ground-truth benchmark: every model runs the same real-world samples across five vision tasks, and answers are scored against ground truth.

Evals updated November 19, 2026Pricing updated November 22, 2026

Strengths and weaknesses

Best showing

TaskScoreMedianRank
Object Detection47.5%57.5%#48 of 73

Weak spot

TaskScoreMedianRank
Identification79.2%86.5%#62 of 73

Still above the median. Most models score well on this task, so the ranks are tight.

Score by taskScoreRank
Object Detection
47.5%#48
Reasoning
52.8%#48
Counting
51.4%#56
Identification
79.2%#62

Gemma 4 31BField median

Ranks out of 73 models

Results by task

TaskScoreField (0 to 100)RankCost / sampleSpeed
Object Detection
47.5%
±0.6, Mean of 3 runs, range 46.8 to 48.0
#48 of 73$0.001155.96s
Counting
51.4%
±2.7, Mean of 3 runs, range 48.6 to 54.0
#56 of 73$0.000313.55s
Identification
79.2%
±1.6, Mean of 3 runs, range 78.1 to 81.3
#62 of 73$0.000212.96s
Reasoning
52.8%
±1.3, Mean of 3 runs, range 51.7 to 54.3
#48 of 73$0.000521.97s

Price vs. performance

Score vs. cost

Overall benchmark score against estimated cost per sample, on a log scale. Upper-left is the sweet spot: high quality at low cost.

72 of 73 models plotted (the rest have no measured figure for this axis) · scores and efficiency pooled across all five tasks at low effort · Gemma 4 31B highlighted

Gemma 4 31B scores are the mean of 3 runs per task at low effort; this model has no separate high-effort tier because its chat template only switches thinking on or off · Methodology

View all Vision Evals →

Self-hosted benchmarks

3 quantizations of Gemma 4 31B, served with vLLM from the published weights and scored on the same tasks as the hosted models. Expand a row for the GPUs it was measured on.

QuantizationWeightsOverall scoreDetection mAP50Single-stream tok/s
FP8Ranked
33.3 GB
57.7%
47.5%63.3
62.5 GB
57.7%
48.2%39.5
23.3 GB
57.6%
48.2%86.6

Rows are ordered by overall score. Every score is the mean of three runs per task under the benchmark protocol (a row marked single run has one), and the gap between quantizations is usually inside the run-to-run range, so read the order as a guide rather than a verdict. Self-hosted rows run with thinking on, the same setting as the hosted frontier models; some hosted open-weight rows still run without thinking, so a self-hosted quant can score above its own hosted API. Quantization still costs a little precision.

Gemma 4 31B Overview

Gemma 4 31B is the largest dense model in Google's Gemma 4 family, built from the same research as Gemini 3 and released as open weights under the Apache 2.0 license. It supports a 256K token context window with text and image input, configurable thinking mode for step-by-step reasoning, and multilingual support across 140+ languages. The unquantized model fits on a single 80GB GPU.

Gemma 4 31B Pricing

Gemma 4 31B costs $0.200 per 1M input tokens and $0.400 per 1M output tokens.

Input$0.200 / 1M tokens
Output$0.400 / 1M tokens

Pricing updated Nov 22, 2026

Deploy Gemma 4 31B with an API

Gemma 4 31B runs as a hosted REST endpoint through Roboflow Workflows. Pick a task, then hand the prompt to your coding agent or copy the code. Deploying the workflow into a free Roboflow workspace replaces the your-workspace and YOUR_API_KEY placeholders with your own.

Connect your agent to Roboflow (once)

Add the Roboflow MCP server

claude mcp add --transport http roboflow https://mcp.roboflow.com/mcp

Run /mcp and authorize Roboflow in your browser when the OAuth flow opens.

Start a new Claude Code session so the MCP loads, then paste the prompt below (it works the same in any agent).

Deploy this workflow to your Roboflow workspace to use it.

Integrate the Roboflow "Gemma 4 31B" workflow into my app.

- Endpoint: POST https://serverless.roboflow.com/<your-workspace>/workflows/gemma-4-31b-captioning
- Auth: send my Roboflow API key as `api_key` in the request body, read from the ROBOFLOW_API_KEY env var (never hardcode).
- Body: { "api_key": ..., "inputs": { `image`: { type: "url" | "base64", value } } }.
- Billing: this workflow needs no provider API key — inference runs on my Roboflow credits. A BYO provider key can be added to the model step in the Roboflow workflow editor later.

With the Roboflow MCP connected, call `workflows_get` on "gemma-4-31b-captioning" to read the exact input schema (the source of truth), then `workflows_run` on a sample image to confirm the output shape before writing code (the MCP is authenticated, so this needs no key). Without the MCP, use the contract above.

Before running the app, set up these keys so it does not error at runtime:
- `ROBOFLOW_API_KEY` (sent as `api_key`) from https://app.roboflow.com/settings/api
Create a .gitignore'd .env with these variables, using placeholder values for any I haven't given you. Then pause and tell me directly, in your reply: the full path to the .env file, exactly which keys I need to paste in, and the link to get each one. Wait for me to confirm I've added them before you run anything. Do not run the app until I confirm.

Then add the integration to my codebase: match my project's language, framework, and conventions; read every key from environment variables (never hardcode); add basic error handling; and include a small runnable example. If you can't tell what language my project uses, ask me.
Installpip install inference-sdk

Deploy this workflow to your Roboflow workspace to use it.

# Inference runs on your Roboflow credits — no provider API key needed. To bill your own provider account instead, add an api_key to the model step in the Roboflow workflow editor.
# 1. Import the library
from inference_sdk import InferenceHTTPClient

# 2. Connect to your workflow
client = InferenceHTTPClient(
  api_url="https://serverless.roboflow.com",
  api_key="YOUR_API_KEY"
)

# 3. Run your workflow on an image
result = client.run_workflow(
  workspace_name="your-workspace",
  workflow_id="gemma-4-31b-captioning",
  images={
    "image": "YOUR_IMAGE.jpg"  # Path to your image file
  },
  use_cache=True  # cache workflow definition for 15 minutes
)

# 4. Get your results
print(result)

Deploy this workflow to your Roboflow workspace to use it.

// Inference runs on your Roboflow credits — no provider API key needed. To bill your own provider account instead, add an api_key to the model step in the Roboflow workflow editor.
const response = await fetch('https://serverless.roboflow.com/your-workspace/workflows/gemma-4-31b-captioning', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    api_key: 'YOUR_API_KEY',
    inputs: {
      "image": {"type": "url", "value": "IMAGE_URL"}
    }
  })
});

const result = await response.json();
console.log(result);

Deploy this workflow to your Roboflow workspace to use it.

# Inference runs on your Roboflow credits — no provider API key needed. To bill your own provider account instead, add an api_key to the model step in the Roboflow workflow editor.
curl --location 'https://serverless.roboflow.com/your-workspace/workflows/gemma-4-31b-captioning' \
--header 'Content-Type: application/json' \
--data '{
  "api_key": "YOUR_API_KEY",
  "inputs": {
    "image": {"type": "url", "value": "IMAGE_URL"}
  }
}'

Gemma 4 31B License

Apache-2.0 · Permissive license

Gemma 4 31B is released under the Apache 2.0 license, so it's commercially usable out of the box: you can ship it in closed-source, commercial products, including self-hosted deployments, without paying a license fee or open-sourcing your code. The Gemma 4 31B license is permissive, so no separate commercial license is required.

Commercial use
Permitted. Because Apache-2.0 is permissive, Gemma 4 31B can ship inside paid products and internal systems with no commercial license and no revenue threshold.
Modification
Permitted. Fine-tuning, quantizing, and distilling are all allowed, and your code changes can stay closed. Files you change must be marked as changed.
Redistribution
Permitted with attribution. Ship the Apache-2.0 license text and any NOTICE file alongside the weights or derived code.

Apache-2.0 grants an express patent license that terminates if you bring a patent claim over the work, and it disclaims warranties. Validate Gemma 4 31B on your own data before you depend on it in production.

Read the full Apache 2.0 license ↗

Do I need a commercial license for Gemma 4 31B?

This is the straightforward case: a permissive license is the best technical solution and you are free to deploy Gemma 4 31B commercially without open-sourcing your own code.

Do not hesitate to reach out with questions for your commercial project. Our team will help you start solving business problems on the first call. See Roboflow commercial licensing for the models included in each plan.

Talk to sales

This model is released under the Apache License 2.0, a permissive open-source license that allows commercial use, modification, distribution, and patent use.

Yes. Under the terms of the Apache 2.0 license, you can freely use this model for commercial purposes, including in proprietary products. You must retain the copyright notice and disclaimers when redistributing.

License information is provided as a guide and is not legal advice.

Gemma 4 31B Details

Details

Categoryopen
ModalityMultimodal
Context Window256K
Parameters31B
LicenseApache-2.0
Release DateApr 2026

Vision Tasks

CaptioningChart Question AnsweringClassificationDocument Question AnsweringImage TaggingMulti-Label ClassificationOCRObject DetectionVision LanguageVisual Question Answering

Usage

Past 30 Days

Alternatives to Gemma 4 31B

Other models worth comparing for similar use cases.

Gemma 4 26B A4B is the Mixture-of-Experts variant in Google's Gemma 4 family, with 25.2B total parameters but only 3.8B active per token.
Qwen3.6-35B-A3B is a sparse Mixture-of-Experts (MoE) multimodal language model developed by the Qwen team at Alibaba Group.
Qwen3.6-27B is a dense 27-billion-parameter multimodal language model developed by Alibaba's Qwen team and released on April 22, 2026.
Gemini 3.5 Flash is a multimodal language model developed by Google DeepMind and released at Google I/O 2026.
Mistral Small 3.1 24B, released on March 17, 2025, is an open-weight multimodal model from Mistral AI, distributed under the Apache-2.0 license.

Frequently Asked Questions About Gemma 4 31B Vision

Yes. Gemma 4 31B accepts image input and handles OCR, object counting, identification, visual reasoning, and object detection. On Roboflow's Vision Evals its strongest task is Object Detection at 47.5% (#48 of 73 at low effort). You can test it on your own image in the demo above.

Not its strength. On Vision Evals, Gemma 4 31B scores 47.5% mAP@50 on object detection (#48 of 73 at low effort) and 51.4% judge-graded accuracy on object counting. For production counting or precise localization, pairing it with a specialized detector like RF-DETR or your own trained model in a Roboflow Workflow is usually more reliable: detect the objects, then count the detections.

On our benchmark's task mix, Gemma 4 31B averages $0.0007 per sample at $0.20 per 1M input and $0.40 per 1M output tokens (#22 of 73 on cost), with an average speed of 34.4s per sample across the benchmark. Actual cost depends on your images and prompts.

On the overall Vision Evals ranking, Gemma 4 31B sits #49 of 73 at 57.7%, just behind Claude Opus 4.8 (58.3%) and just ahead of Gemini 2.5 Pro (57.6%). See the full side-by-side: Gemma 4 31B vs Claude Opus 4.8.