GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks. It provides the same instruction-following and safety-tuning benefits as GPT-5, but with reduced latency and cost. GPT-5 Mini is the successor to OpenAI's o4-mini model.
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).
BYOK only | $0.125 | $1.00 | $0.0125 | -- | -- |
|---|
Input $0.125 / Output $1.00-- tok/s-- uptimeBYOK only
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.
GPT-5 Mini (batch) — Price history
| $0.125 | $1.00 | $0.0125 | — |
- Input $0.125 / Output $1.00
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.
Not enough uptime data to display yet.
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.
Benchmarks
Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.
| Source | Benchmark | Score |
|---|---|---|
| Artificial Analysis | GPT-5 mini (Minimal) GPQA Diamond | 68.7% |
| Artificial Analysis | GPT-5 mini (Minimal) HLE | 5.1% |
| Artificial Analysis | GPT-5 mini (Minimal) IFBench | 45.6% |
| Artificial Analysis | GPT-5 mini (Minimal) τ²-Bench Telecom | 31.9% |
| Artificial Analysis | GPT-5 mini (Minimal) AA-LCR | 39.3% |
| Artificial Analysis | GPT-5 mini (Minimal) CritPt | 0.0% |
| Artificial Analysis | GPT-5 mini (Minimal) LiveCodeBench | 54.5% |
| Artificial Analysis | GPT-5 mini (Minimal) Terminal-Bench Hard | 14.4% |
| Artificial Analysis | GPT-5 mini (Minimal) MMLU-Pro | 77.5% |
| Artificial Analysis | GPT-5 mini (Minimal) AIME 2025 | 46.7% |
| Artificial Analysis | GPT-5 mini (Medium) GPQA Diamond | 80.3% |
| Artificial Analysis | GPT-5 mini (Medium) HLE | 15.9% |
| Artificial Analysis | GPT-5 mini (Medium) IFBench | 71.2% |
| Artificial Analysis | GPT-5 mini (Medium) τ²-Bench Telecom | 71.1% |
| Artificial Analysis | GPT-5 mini (Medium) AA-LCR | 72.3% |
| Artificial Analysis | GPT-5 mini (Medium) CritPt | 1.4% |
| Artificial Analysis | GPT-5 mini (Medium) LiveCodeBench | 69.2% |
| Artificial Analysis | GPT-5 mini (Medium) Terminal-Bench Hard | 28.8% |
| Artificial Analysis | GPT-5 mini (Medium) MMLU-Pro | 82.8% |
| Artificial Analysis | GPT-5 mini (Medium) AIME 2025 | 85.0% |
| Artificial Analysis | GPT-5 mini (High) Coding Index | 15.6 |
| Artificial Analysis | GPT-5 mini (High) Agentic Index | 6.8 |
| Artificial Analysis | GPT-5 mini (High) GPQA Diamond | 82.8% |
| Artificial Analysis | GPT-5 mini (High) HLE | 21.5% |
| Artificial Analysis | GPT-5 mini (High) IFBench | 75.4% |
| Artificial Analysis | GPT-5 mini (High) τ²-Bench Telecom | 68.4% |
| Artificial Analysis | GPT-5 mini (High) AA-LCR | 72.3% |
| Artificial Analysis | GPT-5 mini (High) τ-Bench Banking | 15.5% |
| Artificial Analysis | GPT-5 mini (High) GDPval-AA | 13.7% |
| Artificial Analysis | GPT-5 mini (High) CritPt | 0.0% |
| Artificial Analysis | GPT-5 mini (High) LiveCodeBench | 83.8% |
| Artificial Analysis | GPT-5 mini (High) SciCode | 39.0% |
| Artificial Analysis | GPT-5 mini (High) Terminal-Bench Hard | 33.3% |
| Artificial Analysis | GPT-5 mini (High) Terminal-Bench 4.0 | 0.0% |
| Artificial Analysis | GPT-5 mini (High) Terminal-Bench 2.1 | 3.7% |
| Artificial Analysis | GPT-5 mini (High) MMLU-Pro | 83.7% |
| Artificial Analysis | GPT-5 mini (High) AIME 2025 | 90.7% |
| Design Arena | GPT-5 mini (Default) Models Arena 3D Elo | 1016 |
| Design Arena | GPT-5 mini (Default) Models Arena Asciiart Elo | 1119 |
| Design Arena | GPT-5 mini (Default) Models Arena Code Categories Elo | 1104 |
| Design Arena | GPT-5 mini (Default) Models Arena Data Visualization Elo | 1127 |
| Design Arena | GPT-5 mini (Default) Models Arena Game Development Elo | 1119 |
| Design Arena | GPT-5 mini (Default) Models Arena SVG Elo | 1080 |
| Design Arena | GPT-5 mini (Default) Models Arena UI Component Elo | 1096 |
| Design Arena | GPT-5 mini (Default) Models Arena Website Elo | 1117 |
| Vals AI | GPT 5 Mini AIME | 91.5% |
| Vals AI | GPT 5 Mini CaseLaw v2 | 68.5% |
| Vals AI | GPT 5 Mini CorpFin v2 | 60.2% |
| Vals AI | GPT 5 Mini Finance Agent v1.1 | 51.9% |
| Vals AI | GPT 5 Mini GPQA Diamond | 80.3% |
| Vals AI | GPT 5 Mini IOI v1 | 6.8% |
| Vals AI | GPT 5 Mini LiveCodeBench | 86.6% |
| Vals AI | GPT 5 Mini LegalBench | 81.8% |
| Vals AI | GPT 5 Mini MATH 500 | 94.8% |
| Vals AI | GPT 5 Mini MedCode | 43.0% |
| Vals AI | GPT 5 Mini MedQA | 96.1% |
| Vals AI | GPT 5 Mini MedScribe | 80.6% |
| Vals AI | GPT 5 Mini MGSM | 92.6% |
| Vals AI | GPT 5 Mini MMLU Pro | 82.2% |
| Vals AI | GPT 5 Mini MMMU Pro | 78.9% |
| Vals AI | GPT 5 Mini MortgageTax | 66.9% |
| Vals AI | GPT 5 Mini ProofBench v1 | 9.0% |
| Vals AI | GPT 5 Mini SAGE v1 | 43.0% |
| Vals AI | GPT 5 Mini SWE-bench | 60.8% |
| Vals AI | GPT 5 Mini TaxEval v2 | 75.2% |
| Vals AI | GPT 5 Mini Terminal-Bench 2.0 | 27.0% |
| Vals AI | GPT 5 Mini Vals Index | 44.5% |
| Vals AI | GPT 5 Mini Vals Multimodal Index | 46.2% |
| Vals AI | GPT 5 Mini Vibe Code Bench v1.1 | 14.2% |
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.
Not enough data for apps using this model yet.
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
Documentation
Guides, cookbooks, blog posts, and tools for building with OpenAI: GPT-5 Mini (batch) on OpenRouter.
Frequently asked questions
GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks. It provides the same instruction-following and safety-tuning benefits as GPT-5, but with reduced latency and cost. GPT-5 Mini is the successor to OpenAI's o4-mini model.
The GPT-5 Mini (batch) documentation on OpenRouter includes the GPT-5 is now live blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.
GPT-5 Mini (batch) costs $0.125/M input tokens and $1.00/M output tokens, with separate rates for Cache Read at $0.0125/M tokens and Web Search at $10.00/1K calls.
GPT-5 Mini (batch) has a 400,000 token context window. It supports up to 128,000 completion tokens.
Yes. GPT-5 Mini (batch) accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
GPT-5 Mini (batch) accepts text, images, and files such as PDFs as input and returns text.
GPT-5 Mini (batch) is served by 1 provider on OpenRouter: OpenAI.
GPT-6.1 Terra Pro, GPT-6.1 Terra, GPT-6 Luna Decisions and 58 more are other text models from OpenAI.
GPT-5 Mini (batch) was released on August 7, 2025. Its knowledge cutoff is May 31, 2024.