ModelPriceWatch.com
Last scan 2026-08-09 Models tracked 198 Providers 30 Cheapest paid Granite 4.0 Micro $0.017/Mtok in Every price links to its source

o4-mini

by OpenAI

Deprecated reasoning mid tier
Deprecated — shuts down Oct 23, 2026. Still purchasable until then, but the provider has announced end-of-life. Don't start new work on it. Full deprecation status & replacements →
Availability: Migration target: gpt-5.6-terra.

Today's price · per 1M tokens

Input

$1.10

per 1M tokens

Output

$4.40

per 1M tokens

Blended

$1.93

blended $/1M — 3:1 weighted input:output

Cached input

$0.275

25% of input — prompt caching

Source: official OpenAI pricing · read Jul 4, 2026 MODELPRICEWATCH.COM · 2026-08-09

Overview

Reasoning model with extended thinking. Cost-effective for complex problem solving.

Capabilities

struck through = not supported
Input 2/5
Text ✓ Image ✓ Audio Video PDF
Output 1/5
Text ✓ Image Audio Video Embedding
Features 4/9
Prompt caching ✓ Reasoning ✓ Coding Fast inference Long context ✓ Open weights Multimodal ✓ Web search Realtime

Benchmark performance

accuracy % · higher is better
Avg benchmark score69.2%
Perf per $/Mtok35.9
GPQA Diamond
81.4%
SWE-Bench Verified
68%
Humanity's Last Exam
14.28%
ARC-AGI 2
40%
AIME 2025
92.7%
MMMLU
86%
BFCL
82%
HumanEval
68.1%
MATH 500
90%
Independent composite scores
  • LMSYS Chatbot Arena: 1390 ELO (Human preference)
How it stacks up
  • Ranks #35 of 67 benchmarked models by average score
  • Ranks #44 of 150 comparably-measured models by percentile score, across 10 independent measurements
  • Ranks #85 of 150 comparably-tested models by normalized performance per dollar
  • Strongest at BFCL — 82%, #2 of 23

135 tokens/sec output 35.3s latency to first token (TTFT)

Source: Vellum LLM Leaderboard · updated Aug 9, 2026 · See full rankings →

Specifications

Provider
OpenAI
Context window
200K tokens
Modality
text, image
Parameters
Proprietary
Open source
No — proprietary
Released
Apr 17, 2025
Status
Deprecated ends Oct 23, 2026
Last updated
Jul 4, 2026
Tags
reasoningmultimodal

Availability verified: Jul 25, 2026 — per OpenAI's own deprecation notice