GPT-5.4 Mini: token counter & pricing
OpenAI · exact (uses official tokenizer) · pricing as of 2026-05-31.
- Provider
- OpenAI
- API model ID
gpt-5.4-mini- Context window
- 400,000 tokens
- Input price
- $0.75 per 1M tokens
- Output price
- $4.50 per 1M tokens
- Tokenizer accuracy
- exact (uses official tokenizer)
- Pricing as of
- 2026-05-31
Open the counter to count tokens for GPT-5.4 Mini in real time.
What is GPT-5.4 Mini?
GPT-5.4 Mini is the mid-tier in the GPT-5.4 family, substantially cheaper than full GPT-5.4 ($2.50 input → $0.75 input, 3.3× cheaper) while keeping the same o200k_base tokenizer and 400K context.
The sweet spot between GPT-5.4 Nano (too small for harder tasks) and GPT-5.4 (overkill for routine work).
How tokens are counted here
OpenAI's o200k_base tokenizer. Browser-side via js-tiktoken. Exact.
Pricing notes
$0.75 input / $4.50 output per 1M. Cached input $0.075/M.
No long-context tier, single price across the full 400K window. That makes Mini predictable on cost regardless of prompt length.
For a 1,000-token prompt with 200-token reply: $0.00165 per call, ~$1,650 per 1M calls.
When to use GPT-5.4 Mini
- Default for most cost-conscious production workloads. Where GPT-5.4 is overkill, Mini covers the same use cases at a third of the cost.
- High-volume chat, RAG, extraction, summarization that needs better quality than the Nano tier.
- Single-tier long-context predictability, no surcharge above 128K means you can quote fixed budgets to stakeholders.
When not to use it:
- Multi-step reasoning where errors compound. GPT-5.4 or 5.5.
- Bulk classification at maximum cost compression. GPT-5.4 Nano at $0.20/$1.25 is 3× cheaper still.
- Anywhere Claude Sonnet has measurably won on your task.
Common questions
How does GPT-5.4 Mini compare to GPT-5 Mini?
GPT-5 Mini: $0.25/$2, 3× cheaper on input than 5.4 Mini ($0.75/$4.50). The price gap reflects the reasoning improvement: 5.4 Mini benchmarks meaningfully better on multi-step tasks. For routine workloads GPT-5 Mini is the rational default; reach for 5.4 Mini when you've measured a quality win.
What about Gemini 3 Flash?
Gemini 3 Flash ($0.50/$3) is cheaper on both input (-33%) and output (-33%) than 5.4 Mini, with 1M context and multimodal input. The trade-off is OpenAI's more mature tool-use reliability and function-calling. Test both with your prompts.
Caching savings?
$0.075/M cached input is 10% of standard. Worth it on agent loops with stable system prompts.
Compare GPT-5.4 Mini to other models
- GPT-5.5 (OpenAI, $5.00/$30.00)
- GPT-5.5 Pro (OpenAI, $30.00/$180.00)
- GPT-5.4 (OpenAI, $2.50/$15.00)
- GPT-5.4 Nano (OpenAI, $0.20/$1.25)
- GPT-5.4 Pro (OpenAI, $30.00/$180.00)
- GPT-5.3 (OpenAI, $1.75/$14.00)
- GPT-5.2 (OpenAI, $1.75/$14.00)
- GPT-5.2 Pro (OpenAI, $21.00/$168.00)
- GPT-5.1 (OpenAI, $1.25/$10.00)
- GPT-5 (OpenAI, $1.25/$10.00)
- GPT-5 Mini (OpenAI, $0.25/$2.00)
- GPT-5 Nano (OpenAI, $0.05/$0.40)
- GPT-5 Pro (OpenAI, $15.00/$120.00)
- GPT-4.1 (OpenAI, $2.00/$8.00)
- GPT-4.1 Mini (OpenAI, $0.40/$1.60)
- GPT-4.1 Nano (OpenAI, $0.10/$0.40)
- o3 (OpenAI, $2.00/$8.00)
- o3-mini (OpenAI, $1.10/$4.40)
- o3-pro (OpenAI, $20.00/$80.00)
- o4-mini (OpenAI, $1.10/$4.40)
- GPT-4o (OpenAI, $2.50/$10.00)
- GPT-4o mini (OpenAI, $0.15/$0.60)
- GPT-4 Turbo (OpenAI, $10.00/$30.00)
- Qwen 2.5 Coder 32B (Alibaba, $0.80/$0.80)
- Llama 3.3 70B (Meta, $0.88/$0.88)
- DeepSeek V3.1 (DeepSeek, $0.60/$1.70)