OpenAI · Efficient
GPT 5 Mini
A lightweight version of GPT-5 for cost-sensitive applications. Balances advanced capabilities with efficient resource usage.
- Reasoning
- Vision
- Input
- Text, Images
- Output
- Text
- Reasoning effort
- minimal · low · medium · high
- Discovery tier
- Efficient
Pricing
USD per 1M tokens- Input
- $0.25
- Output
- $2.00
- Cache read
- $0.025
Reference prices as published by the model provider. API example
Specifications
- Context window
- 400K
- Max output
- 128K
- Average output (7 days)
- 891 tokens
- Longer than 67% of 465 models
Side by side
GPT 5 Mini compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| GPT 5 MiniOpenAIThis model | DeepSeek V3.2 Exp ThinkingDeepSeekRemove | |
|---|---|---|
| Provider | OpenAI | DeepSeek |
| Discovery tier | Efficient | Open weights |
| Added | Aug 7, 2025 | Jan 1, 2024 |
| Limits | ||
| Context window | 400KBest | 163.84K |
| Max output | 128KBest | 65.54K |
| Average output | 891 tokens | 917 tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.25Best | $0.28 |
| Output | $2.00 | $0.42Best |
| Cache read | $0.025 | — |
| Performance | ||
| Throughput | — | 11.3 tok/s |
| Time to first token | — | 5.2 s |
| Request success (24h) | 100.0% | 100.0% |
| Benchmark indexes | ||
| Intelligence Index | 16.8Best | 16.6 |
| Coding Index | 15.6 | — |
| Math Index | 90.7Best | 87.7 |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Vision | Yes | No |
| Reasoning effort | minimal, low, medium, high | — |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- GPT 5 Mini
- DeepSeek V3.2 Exp Thinking
Intelligence Index
- GPT 5 Mini16.8
- DeepSeek V3.2 Exp Thinking16.6
GPT 5 Mini tested at high reasoning effort
GPT 5 Mini: Better than 45% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- GPT 5 Mini: 15.6
- DeepSeek V3.2 Exp Thinking: —
Math Index
Composite of the math evaluations
- GPT 5 Mini: 90.7Top
- DeepSeek V3.2 Exp Thinking: 87.7
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- GPT 5 Mini: 6.5%
- DeepSeek V3.2 Exp Thinking: —
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- GPT 5 Mini: 0.9%
- DeepSeek V3.2 Exp Thinking: —
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- GPT 5 Mini: 428 Elo
- DeepSeek V3.2 Exp Thinking: —
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- GPT 5 Mini: 766 Elo
- DeepSeek V3.2 Exp Thinking: —
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- GPT 5 Mini: 8.4%
- DeepSeek V3.2 Exp Thinking: —
AA-LCRv1.1
Reasoning across long documents
- GPT 5 Mini: 72.3%Top
- DeepSeek V3.2 Exp Thinking: 72.3%Top
MLCR-AA
Long-context reasoning over medical records
- GPT 5 Mini: 8.9%
- DeepSeek V3.2 Exp Thinking: —
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- GPT 5 Mini: 21.5%Top
- DeepSeek V3.2 Exp Thinking: 14.9%
GPQA Diamond
Graduate-level science questions
- GPT 5 Mini: 82.8%Top
- DeepSeek V3.2 Exp Thinking: 79.7%
MMLU-Pro
Multi-subject knowledge and reasoning
- GPT 5 Mini: 83.7%
- DeepSeek V3.2 Exp Thinking: 85.0%Top
IFBench
Following precise output instructions
- GPT 5 Mini: 75.4%Top
- DeepSeek V3.2 Exp Thinking: 54.1%
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- GPT 5 Mini: 0.0%
- DeepSeek V3.2 Exp Thinking: —
SciCode
Scientific computing in Python
- GPT 5 Mini: 39.0%
- DeepSeek V3.2 Exp Thinking: —
LiveCodeBench
Recent competitive programming problems
- GPT 5 Mini: 83.8%Top
- DeepSeek V3.2 Exp Thinking: 78.9%
Math
AIME 2025
Competition mathematics, 2025 problems
- GPT 5 Mini: 90.7%Top
- DeepSeek V3.2 Exp Thinking: 87.7%
Legacy evaluations
Terminal-Bench Hard
Earlier terminal task suite
- GPT 5 Mini: 33.3%Top
- DeepSeek V3.2 Exp Thinking: 31.1%
τ²-Bench Telecom
Tool use in a customer-support simulation
- GPT 5 Mini: 68.4%Top
- DeepSeek V3.2 Exp Thinking: 33.9%
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- GPT 5 Mini: 72.3%Top
- DeepSeek V3.2 Exp Thinking: 72.3%Top
Scores from Artificial Analysis, updated Oct 26, 2026. GPT 5 Mini was tested at high reasoning effort; compare scores measured at the same effort.Artificial Analysis
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Prompt cache hits
- 22.1%
- About $0.20 per 1M input tokens with caching
- Request success (24h)
- 100.0%
- Reliable
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
- Healthy (98% or more)
- Degraded (90–98%)
- Down (under 90%)
- No requests
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-5-mini",
"reasoning_effort": "medium",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="openai/gpt-5-mini",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)OpenAI
Related models
Other OpenAI models, starting with the same family.