Roboflow

OpenAIOpenAIReleased Mar 2026Proprietary

GPT-5.4 Mini

GPT-5.4 Mini scores 53.2% overall on Roboflow's Vision Evals (November 19, 2026 snapshot), #61 of 73 models (GPT-5.6 Terra ranks higher at 66.0%).

Its weakest task is object detection (15.8%, #71 of 73). It is a proprietary model from OpenAI, so commercial use follows OpenAI's terms.

Vision Evals53.2%Field median 65.7% · #61 of 73
Cost$0.0034/ sampleField median $0.0040 · #35 of 73
Speed5.5s/ sampleField median 10.2s · #5 of 73

GPT-5.4 Mini Interactive Demo

Model settings

Thinking level

Max output tokens

Default 65,536 · max 65,536

Sign in to adjust thinking and output length per run.

Results appear here. Add an image or pick an example to run GPT-5.4 Mini.

GPT-5.4 Mini 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
Reasoning57.0%61.1%#44 of 73

Weak spot

TaskScoreMedianRank
Object Detection15.8%57.5%#71 of 73

Below the field median.

Score by taskScoreRank
Reasoning
57.0%#44
Counting
58.6%#46
Identification
83.3%#47
OCR
51.1%#53
Object Detection
15.8%#71

GPT-5.4 MiniField median

Ranks out of 73 models (57 for OCR)

Results by task

TaskScoreField (0 to 100)RankCost / sampleSpeed
Object Detection (low)
15.8%
±0.4, Mean of 3 runs, range 15.3 to 16.1
#71 of 73$0.00447.32s
Object Detection (high)
16.6%
±0.8, Mean of 3 runs, range 15.8 to 17.4
#41 of 41$0.03042.45s
Counting (low)
58.6%
±2.0, Mean of 3 runs, range 56.8 to 60.8
#46 of 73$0.00193.84s
Counting (high)
64.9%
±2.0, Mean of 3 runs, range 63.5 to 67.6
#34 of 41$0.007311.15s
Identification (low)
83.3%
±1.6, Mean of 3 runs, range 81.3 to 84.4
#47 of 73$0.00132.61s
Identification (high)
82.3%
±3.1, Mean of 3 runs, range 78.1 to 84.4
#37 of 41$0.00559.04s
OCR (low)
51.1%
#53 of 57$0.00355.44s
OCR (high)
55.8%
#49 of 57$0.02733.34s
Reasoning (low)
57.0%
±3.3, Mean of 3 runs, range 54.3 to 60.9
#44 of 73$0.00224.18s
Reasoning (high)
64.0%
±1.3, Mean of 3 runs, range 62.9 to 65.6
#43 of 60$0.009613.47s
  • Thinking longer helps: 0.8 points higher on object detection at high effort for 6.8x the cost and 5.8x the latency.
  • Thinking longer helps: 6.3 points higher on counting at high effort for 3.8x the cost and 2.9x the latency.
  • Thinking longer does not help: 1 points lower on identification at high effort for 4.1x the cost and 3.5x the latency.
  • Thinking longer helps: 4.6 points higher on ocr at high effort for 7.5x the cost and 6.1x the latency.
  • Thinking longer helps: 7.1 points higher on reasoning at high effort for 4.3x the cost and 3.2x the latency.

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 · GPT-5.4 mini highlighted

GPT-5.4 Mini scores are the mean of 3 runs per task at both low and high effort; OCR is a single preliminary run · Methodology

View all Vision Evals →

GPT-5.4 Mini Overview

GPT-5.4 mini is a fast, cost-efficient model developed by OpenAI and released on March 17, 2026, optimized for high-throughput workloads and subagent orchestration. It supports text and image inputs within a 400,000-token context window, making it ideal for processing extensive visual datasets and large codebases in a single request. Designed for low-latency production environments, the model integrates with key API features including function calling, web search, and tool-based computer use, allowing it to assist in automated workflows that require navigating digital interfaces.

GPT-5.4 Mini Pricing

GPT-5.4 Mini costs $0.750 per 1M input tokens and $4.50 per 1M output tokens.

Input$0.750 / 1M tokens
Output$4.50 / 1M tokens
Cached input$0.075 / 1M tokens

Pricing updated Nov 22, 2026

Deploy GPT-5.4 Mini with an API

GPT-5.4 Mini 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 "GPT-5.4 Mini" workflow into my app.

- Endpoint: POST https://serverless.roboflow.com/<your-workspace>/workflows/gpt-5-4-mini-open-prompt
- 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 }, `prompt`: text } }.
- 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 "gpt-5-4-mini-open-prompt" 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="gpt-5-4-mini-open-prompt",
  images={
    "image": "YOUR_IMAGE.jpg"  # Path to your image file
  },
  parameters={
    "prompt": "Describe what you see in the image"
  },
  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/gpt-5-4-mini-open-prompt', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    api_key: 'YOUR_API_KEY',
    inputs: {
      "image": {"type": "url", "value": "IMAGE_URL"},
      "prompt": "Describe what you see in the image"
    }
  })
});

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/gpt-5-4-mini-open-prompt' \
--header 'Content-Type: application/json' \
--data '{
  "api_key": "YOUR_API_KEY",
  "inputs": {
    "image": {"type": "url", "value": "IMAGE_URL"},
    "prompt": "Describe what you see in the image"
  }
}'

GPT-5.4 Mini License

Proprietary

GPT-5.4 Mini is proprietary: the weights are not distributed, and the GPT-5.4 Mini license is the vendor's commercial terms of service that you accept when you call the API.

Commercial use
Permitted under the vendor terms, typically metered per token or per request, with the vendor usage policy applying to your inputs and outputs.
Modification
Not available. GPT-5.4 Mini weights are closed, so you can configure prompts and use vendor-hosted fine-tuning where it is offered, but you cannot modify the model itself.
Redistribution
Not permitted. You cannot self-host or resell the model; you build on the hosted API instead.

Vendor terms govern data retention, whether your inputs can be trained on, rate limits, and regional availability, and they can change with notice. Review them if you handle regulated or customer data.

Do I need a commercial license for GPT-5.4 Mini?

Proprietary terms are set by the vendor rather than negotiated per project, and no open-source obligation attaches to your code. If you would rather deploy an open model instead of GPT-5.4 Mini, Roboflow's licensing page lists the models whose commercial license is included when you deploy on Roboflow.

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

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

GPT-5.4 Mini Details

Details

Categoryclosed
ModalityMultimodal
Context Window400K
LicenseProprietary
Release DateMar 2026

Vision Tasks

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

Usage

Past 30 Days

Alternatives to GPT-5.4 Mini

Other models worth comparing for similar use cases.

GPT-6.1 Terra is an OpenAI model released on October 26, 2026 for visual reasoning, computer use and writing.
Gemini 3.8 Flash is a natively multimodal reasoning model in Google's Gemini 3 series, positioned as the speed and cost oriented Flash tier while targeting long-horizon software…
Claude Sonnet 5.5 is a proprietary multimodal language model from Anthropic and the second release in the Claude 5.5 family, following Claude Opus 5.5.
Qwen3.8-Flash is a multimodal mixture-of-experts model from the Qwen team at Alibaba, and the production counterpart of the open-weight Qwen3.8-Flash-Next preview that introduces…
GLM-5.3-Flash is the first natively multimodal model in Z.ai's GLM-5 series, a mixture-of-experts transformer with roughly 320 billion total parameters and 18 billion activated…

Other OpenAI GPT Mini models

Other versions in the same family as GPT-5.4 Mini.

Frequently Asked Questions About GPT-5.4 Mini Vision

Yes. GPT-5.4 Mini accepts image input and handles OCR, object counting, identification, visual reasoning, and object detection. On Roboflow's Vision Evals its strongest task is Reasoning at 57% (#44 of 73 at low effort). You can test it on your own image in the demo above.

Yes. it scores 51.1% (#53 of 57 at low effort) on Vision Evals OCR, which covers single readings, full transcriptions, structured JSON, and locating text with boxes. Within that, full transcriptions score 77.3%, and returning the fields of a label or form as JSON scores 73.6%.

Not its strength. On Vision Evals, GPT-5.4 Mini scores 15.8% mAP@50 on object detection (#71 of 73 at low effort) and 58.6% 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, GPT-5.4 Mini averages $0.0034 per sample at $0.75 per 1M input and $4.50 per 1M output tokens (#35 of 73 on cost), with an average speed of 5.5s per sample across the benchmark. Actual cost depends on your images and prompts.

On the overall Vision Evals ranking, GPT-5.4 Mini sits #61 of 73 at 53.2%, just behind Gemma 4 26B-A4B (54.1%) and just ahead of Kimi K2.6 (52.8%). See the full side-by-side: GPT-5.4 Mini vs Gemma 4 26B-A4B.