GPT-5 Mini
OpenAI · moe · 80B parameters · 400,000 context
Parameters
80B
Context Window
391K tokens
Architecture
MoE
Best GPU
H200 SXM
Cheapest API
$2.00/M
Intelligence Brief
GPT-5 Mini is a 80B parameter Mixture-of-Experts (16 experts, 2 active) model from OpenAI, featuring Grouped Query Attention (GQA) with 80 layers and 10,240 hidden dimensions. With a 400,000 token context window, it supports tools, vision, structured output, code, math, multilingual, reasoning. The most cost-effective API deployment is via openai at $2.00/M output tokens. For self-hosted inference, H200 SXM delivers optimal throughput at $5106/month.
Provider pricing
1 provider · canonical: openai| Provider | Input $/M | Output $/M ▲ | Notes |
|---|---|---|---|
| openaicanonical | $0.250 | $2.00 | cheapest input · cheapest output |
Prices update via the nightly pricing cron + admin approvals at /admin/ingest-queue. The leaderboard's Input/Output cells show the canonical rate above; this table shows the full spread.
Recent changes
Loading…
Related models
5 suggestions- GPT-4.5 PreviewGPT · 300B$150.00/M out
- GPT-5.5GPT · 110B$30.00/M out
- GPT-5 NanoGPT · 4B$0.400/M out
- GPT-5GPT · 90B$10.00/M out
- GPT-3.5 TurboGPT-3.5 · 20B$1.50/M out
Picks: same family first, then same vendor within ±2× params, then top tag-overlap matches. Price shown is the cheapest Output $/M across providers — the row's page shows the canonical anchor.
Architecture Details
Memory Requirements
BF16 Weights
160.0 GB
FP8 Weights
80.0 GB
INT4 Weights
40.0 GB
GPU Compatibility Matrix
GPT-5 Mini is compatible with 33% of GPU configurations across 41 GPUs at 3 precision levels.
GPU Recommendations
BF16 · 2 GPUs · tensorrt-llm
95/100
score
Throughput
560.0 tok/s
Latency (ITL)
1.8ms
Est. TTFT
0ms
Cost/Month
$5106
Cost/M Tokens
$3.47
BF16 · 1 GPU · tensorrt-llm
93/100
score
Throughput
560.0 tok/s
Latency (ITL)
1.8ms
Est. TTFT
0ms
Cost/Month
$4271
Cost/M Tokens
$2.90
BF16 · 1 GPU · tensorrt-llm
93/100
score
Throughput
560.0 tok/s
Latency (ITL)
1.8ms
Est. TTFT
0ms
Cost/Month
$6169
Cost/M Tokens
$4.19
Deployment Options
API Deployment
openai
$2.00/M
output tokens
Single GPU
B100 SXM
$4271/mo
Min VRAM: 80 GB
Multi-GPU
H200 SXM x2
560.0 tok/s
TP· $5106/mo
API Pricing Comparison
| Provider | Input $/M | Output $/M | Badges |
|---|---|---|---|
| openai | $0.25 | $2.00 | Cheapest |
Cost Analysis
| Provider | Input $/M | Output $/M | ~Monthly Cost |
|---|---|---|---|
| openaiBest Value | $0.25 | $2.00 | $11 |
Cost per 1,000 Requests
Short (500 tok)
$0.53
via openai
Medium (2K tok)
$2.10
via openai
Long (8K tok)
$6.00
via openai
Performance Estimates
Throughput by GPU
VRAM Breakdown (H200 SXM, BF16)
Capabilities
Features
Supported Frameworks
Supported Precisions
Where to Deploy GPT-5 Mini
Self-Hosted Infrastructure
Similar Models
Qwen 2.5 72B
72.7B params · dense
Quality: 77
from $0.00/M
Qwen 2.5 Math 72B
72.7B params · dense
Quality: 50
from $0.90/M
Qwen 2.5 VL 72B
72.7B params · dense
Quality: 50
from $0.75/M
Molmo 72B
72B params · dense
Quality: 78
Dolphin 2.9 72B
72B params · dense
Quality: 50
Frequently Asked Questions
How much VRAM does GPT-5 Mini need for inference?
GPT-5 Mini requires approximately 160.0 GB of VRAM at BF16 precision, 80.0 GB at FP8, or 40.0 GB at INT4 quantization. Additional VRAM is needed for KV-cache (204800 bytes per token) and activations (~4.00 GB).
What is the best GPU for GPT-5 Mini?
The top recommended GPU for GPT-5 Mini is the H200 SXM (x2) using BF16 precision. It achieves approximately 560.0 tokens/sec at an estimated cost of $5106/month ($3.47/M tokens). Score: 95/100.
How much does GPT-5 Mini inference cost?
GPT-5 Mini API inference starts from $0.25/M input tokens and $2.00/M output tokens. Self-hosted inference costs depend on your GPU configuration — use our ROI calculator for a detailed breakdown.