PE Collective
GPT-4o Mini Pricing: The Cheapest Way to Use OpenAI (April 2026)
GPT-4o Mini is OpenAI's cheapest model in the GPT-4 family at $0.15/$0.60 per million tokens. Launched July 2024, it handles classification, extraction, routing, and simple generation at 94% less than GPT-4o. Now competes with GPT-4.1 Nano ($0.10/$0.40) for the budget tier.
GPT-4o Mini
- ✓ Cheapest GPT-4 family model from OpenAI
- ✓ 128K context window
- ✓ Function calling and structured outputs
- ✓ Vision support for image inputs
GPT-4o Mini (Cached)
- ✓ 50% discount on input tokens with prompt caching
- ✓ Output tokens remain at standard rate
- ✓ Best for apps with repeated system prompts
- ✓ Cache hits automatic — no code changes needed
GPT-4o Mini (Batch)
- ✓ 50% off both input and output tokens
- ✓ 24-hour turnaround SLA
- ✓ Same quality as synchronous API
- ✓ Best for bulk classification and labeling
GPT-4o Mini (Fine-tuned)
- ✓ Double base price for custom fine-tuned models
- ✓ Training costs $0.30 per 1M tokens on top of inference
- ✓ Best for domain-specific classification tasks
- ✓ Requires at least 10 training examples
GPT-4o Mini Pricing Table (April 2026)
E
v
e
r
y
G
P
T
-
4
o
M
i
n
i
p
r
i
c
i
n
g
v
a
r
i
a
n
t
i
n
o
n
e
t
a
b
l
e
.
A
l
l
p
r
i
c
e
s
a
r
e
p
e
r
1
m
i
l
l
i
o
n
t
o
k
e
n
s
.
GPT-4o Mini vs GPT-4.1 Nano: Which Budget Model Wins?
G
P
T
-
4
.
1
N
a
n
o
l
a
u
n
c
h
e
d
a
s
O
p
e
n
A
I
'
s
n
e
w
u
l
t
r
a
-
c
h
e
a
p
o
p
t
i
o
n
a
n
d
d
i
r
e
c
t
l
y
c
o
m
p
e
t
e
s
w
i
t
h
G
P
T
-
4
o
M
i
n
i
.
N
a
n
o
i
s
3
3
%
c
h
e
a
p
e
r
o
n
b
o
t
h
i
n
p
u
t
a
n
d
o
u
t
p
u
t
t
o
k
e
n
s
.
B
o
t
h
m
o
d
e
l
s
t
a
r
g
e
t
t
h
e
s
a
m
e
u
s
e
c
a
s
e
s
:
c
l
a
s
s
i
f
i
c
a
t
i
o
n
,
e
x
t
r
a
c
t
i
o
n
,
r
o
u
t
i
n
g
,
a
n
d
s
i
m
p
l
e
g
e
n
e
r
a
t
i
o
n
.
T
h
e
k
e
y
d
i
f
f
e
r
e
n
c
e
s
a
r
e
c
o
n
t
e
x
t
w
i
n
d
o
w
s
i
z
e
(
N
a
n
o
s
u
p
p
o
r
t
s
1
M
t
o
k
e
n
s
v
e
r
s
u
s
M
i
n
i
'
s
1
2
8
K
)
a
n
d
r
e
c
e
n
c
y
(
N
a
n
o
b
e
n
e
f
i
t
s
f
r
o
m
n
e
w
e
r
t
r
a
i
n
i
n
g
d
a
t
a
a
n
d
a
r
c
h
i
t
e
c
t
u
r
e
i
m
p
r
o
v
e
m
e
n
t
s
)
.
F
o
r
m
o
s
t
s
i
m
p
l
e
t
a
s
k
s
,
N
a
n
o
i
s
t
h
e
b
e
t
t
e
r
d
e
f
a
u
l
t
.
M
i
n
i
s
t
i
l
l
w
i
n
s
i
f
y
o
u
h
a
v
e
f
i
n
e
-
t
u
n
e
d
m
o
d
e
l
s
y
o
u
d
o
n
o
t
w
a
n
t
t
o
r
e
t
r
a
i
n
,
o
r
i
f
y
o
u
r
s
p
e
c
i
f
i
c
t
a
s
k
b
e
n
c
h
m
a
r
k
s
h
i
g
h
e
r
o
n
M
i
n
i
.
GPT-4o Mini vs GPT-4o: When to Upgrade
G
P
T
-
4
o
c
o
s
t
s
$
2
.
5
0
/
$
1
0
.
0
0
p
e
r
m
i
l
l
i
o
n
t
o
k
e
n
s
—
r
o
u
g
h
l
y
1
7
x
m
o
r
e
t
h
a
n
M
i
n
i
o
n
i
n
p
u
t
a
n
d
o
u
t
p
u
t
.
T
h
e
q
u
a
l
i
t
y
g
a
p
m
a
t
t
e
r
s
m
o
s
t
o
n
c
o
m
p
l
e
x
r
e
a
s
o
n
i
n
g
,
n
u
a
n
c
e
d
w
r
i
t
i
n
g
,
m
u
l
t
i
-
s
t
e
p
t
o
o
l
u
s
e
,
a
n
d
l
o
n
g
-
f
o
r
m
c
o
n
t
e
n
t
g
e
n
e
r
a
t
i
o
n
.
M
i
n
i
m
a
t
c
h
e
s
G
P
T
-
4
o
c
l
o
s
e
l
y
o
n
s
i
m
p
l
e
c
l
a
s
s
i
f
i
c
a
t
i
o
n
(
9
0
%
+
a
g
r
e
e
m
e
n
t
o
n
b
i
n
a
r
y
t
a
s
k
s
)
b
u
t
f
a
l
l
s
b
e
h
i
n
d
o
n
a
n
y
t
h
i
n
g
r
e
q
u
i
r
i
n
g
c
h
a
i
n
-
o
f
-
t
h
o
u
g
h
t
r
e
a
s
o
n
i
n
g
o
r
s
u
b
t
l
e
d
i
s
t
i
n
c
t
i
o
n
s
.
A
c
o
m
m
o
n
a
r
c
h
i
t
e
c
t
u
r
e
u
s
e
s
M
i
n
i
f
o
r
i
n
i
t
i
a
l
t
r
i
a
g
e
a
n
d
r
o
u
t
e
s
o
n
l
y
c
o
m
p
l
e
x
r
e
q
u
e
s
t
s
t
o
G
P
T
-
4
o
o
r
G
P
T
-
4
.
1
,
c
u
t
t
i
n
g
c
o
s
t
s
b
y
7
0
-
8
0
%
.
Batch API Pricing for GPT-4o Mini
T
h
e
B
a
t
c
h
A
P
I
p
r
o
c
e
s
s
e
s
r
e
q
u
e
s
t
s
a
s
y
n
c
h
r
o
n
o
u
s
l
y
a
t
5
0
%
o
f
f
b
o
t
h
i
n
p
u
t
a
n
d
o
u
t
p
u
t
t
o
k
e
n
s
.
G
P
T
-
4
o
M
i
n
i
B
a
t
c
h
p
r
i
c
i
n
g
i
s
$
0
.
0
7
5
i
n
p
u
t
a
n
d
$
0
.
3
0
o
u
t
p
u
t
p
e
r
m
i
l
l
i
o
n
t
o
k
e
n
s
.
J
o
b
s
c
o
m
p
l
e
t
e
w
i
t
h
i
n
2
4
h
o
u
r
s
.
S
u
b
m
i
t
a
J
S
O
N
L
f
i
l
e
o
f
C
h
a
t
C
o
m
p
l
e
t
i
o
n
s
r
e
q
u
e
s
t
s
a
n
d
p
o
l
l
f
o
r
r
e
s
u
l
t
s
.
B
e
s
t
f
o
r
:
d
a
t
a
s
e
t
l
a
b
e
l
i
n
g
,
c
o
n
t
e
n
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
,
b
u
l
k
e
x
t
r
a
c
t
i
o
n
,
a
n
d
e
v
a
l
u
a
t
i
o
n
p
i
p
e
l
i
n
e
s
w
h
e
r
e
l
a
t
e
n
c
y
i
s
n
o
t
c
r
i
t
i
c
a
l
.
Fine-Tuning Costs for GPT-4o Mini
F
i
n
e
-
t
u
n
i
n
g
G
P
T
-
4
o
M
i
n
i
c
o
s
t
s
$
0
.
3
0
p
e
r
1
M
t
r
a
i
n
i
n
g
t
o
k
e
n
s
.
O
n
c
e
t
r
a
i
n
e
d
,
i
n
f
e
r
e
n
c
e
c
o
s
t
s
d
o
u
b
l
e
:
$
0
.
3
0
i
n
p
u
t
a
n
d
$
1
.
2
0
o
u
t
p
u
t
p
e
r
m
i
l
l
i
o
n
t
o
k
e
n
s
.
A
t
y
p
i
c
a
l
f
i
n
e
-
t
u
n
i
n
g
j
o
b
o
n
1
0
0
K
t
r
a
i
n
i
n
g
t
o
k
e
n
s
c
o
s
t
s
a
b
o
u
t
$
0
.
0
3
f
o
r
t
r
a
i
n
i
n
g
a
l
o
n
e
.
T
h
e
r
e
a
l
c
o
s
t
i
s
i
n
f
e
r
e
n
c
e
—
i
f
y
o
u
r
u
n
1
M
r
e
q
u
e
s
t
s
p
e
r
m
o
n
t
h
,
t
h
e
2
x
i
n
f
e
r
e
n
c
e
p
r
e
m
i
u
m
a
d
d
s
u
p
f
a
s
t
.
C
o
n
s
i
d
e
r
w
h
e
t
h
e
r
p
r
o
m
p
t
e
n
g
i
n
e
e
r
i
n
g
o
r
f
e
w
-
s
h
o
t
e
x
a
m
p
l
e
s
c
a
n
a
c
h
i
e
v
e
t
h
e
s
a
m
e
q
u
a
l
i
t
y
b
e
f
o
r
e
c
o
m
m
i
t
t
i
n
g
t
o
f
i
n
e
-
t
u
n
i
n
g
.
F
i
n
e
-
t
u
n
e
d
M
i
n
i
m
o
d
e
l
s
s
t
i
l
l
c
a
n
n
o
t
m
a
t
c
h
t
h
e
r
e
a
s
o
n
i
n
g
a
b
i
l
i
t
y
o
f
b
a
s
e
G
P
T
-
4
.
1
o
r
G
P
T
-
4
o
.
Real-World Cost Examples
W
h
a
t
G
P
T
-
4
o
M
i
n
i
a
c
t
u
a
l
l
y
c
o
s
t
s
f
o
r
c
o
m
m
o
n
w
o
r
k
l
o
a
d
s
a
t
d
i
f
f
e
r
e
n
t
s
c
a
l
e
s
.
A
l
l
e
s
t
i
m
a
t
e
s
a
s
s
u
m
e
a
v
e
r
a
g
e
t
o
k
e
n
c
o
u
n
t
s
p
e
r
r
e
q
u
e
s
t
.
Rate Limits for GPT-4o Mini
O
p
e
n
A
I
s
e
t
s
r
a
t
e
l
i
m
i
t
s
b
a
s
e
d
o
n
y
o
u
r
c
u
m
u
l
a
t
i
v
e
A
P
I
s
p
e
n
d
i
n
g
.
G
P
T
-
4
o
M
i
n
i
s
h
a
r
e
s
t
h
e
s
a
m
e
t
i
e
r
s
y
s
t
e
m
a
s
o
t
h
e
r
m
o
d
e
l
s
b
u
t
t
y
p
i
c
a
l
l
y
h
a
s
h
i
g
h
e
r
t
o
k
e
n
-
p
e
r
-
m
i
n
u
t
e
a
l
l
o
w
a
n
c
e
s
d
u
e
t
o
i
t
s
l
o
w
e
r
c
o
s
t
.
N
e
w
a
c
c
o
u
n
t
s
s
t
a
r
t
a
t
t
h
e
f
r
e
e
t
i
e
r
.
When GPT-4o Mini Is NOT Enough
G
P
T
-
4
o
M
i
n
i
f
a
i
l
s
o
r
u
n
d
e
r
p
e
r
f
o
r
m
s
o
n
s
e
v
e
r
a
l
t
a
s
k
c
a
t
e
g
o
r
i
e
s
.
C
o
m
p
l
e
x
m
u
l
t
i
-
s
t
e
p
r
e
a
s
o
n
i
n
g
c
o
n
s
i
s
t
e
n
t
l
y
p
r
o
d
u
c
e
s
e
r
r
o
r
s
t
h
a
t
G
P
T
-
4
.
1
o
r
G
P
T
-
4
o
h
a
n
d
l
e
c
o
r
r
e
c
t
l
y
.
N
u
a
n
c
e
d
w
r
i
t
i
n
g
t
a
s
k
s
(
t
o
n
e
m
a
t
c
h
i
n
g
,
p
e
r
s
u
a
s
i
v
e
c
o
p
y
,
c
r
e
a
t
i
v
e
f
i
c
t
i
o
n
)
s
h
o
w
a
c
l
e
a
r
q
u
a
l
i
t
y
g
a
p
.
L
o
n
g
-
c
o
n
t
e
x
t
r
e
t
r
i
e
v
a
l
d
e
g
r
a
d
e
s
p
a
s
t
3
2
K
t
o
k
e
n
s
e
v
e
n
t
h
o
u
g
h
t
h
e
1
2
8
K
w
i
n
d
o
w
t
e
c
h
n
i
c
a
l
l
y
a
c
c
e
p
t
s
m
o
r
e
.
M
u
l
t
i
-
t
u
r
n
a
g
e
n
t
w
o
r
k
f
l
o
w
s
w
i
t
h
t
o
o
l
u
s
e
p
r
o
d
u
c
e
m
o
r
e
h
a
l
l
u
c
i
n
a
t
e
d
f
u
n
c
t
i
o
n
c
a
l
l
s
.
C
o
d
e
g
e
n
e
r
a
t
i
o
n
f
o
r
n
o
n
-
t
r
i
v
i
a
l
t
a
s
k
s
(
r
e
f
a
c
t
o
r
i
n
g
,
a
r
c
h
i
t
e
c
t
u
r
e
c
h
a
n
g
e
s
,
m
u
l
t
i
-
f
i
l
e
e
d
i
t
s
)
m
i
s
s
e
s
e
d
g
e
c
a
s
e
s
.
I
f
y
o
u
r
a
p
p
l
i
c
a
t
i
o
n
f
a
l
l
s
i
n
t
o
a
n
y
o
f
t
h
e
s
e
c
a
t
e
g
o
r
i
e
s
,
b
u
d
g
e
t
f
o
r
a
b
i
g
g
e
r
m
o
d
e
l
a
n
d
u
s
e
M
i
n
i
o
n
l
y
f
o
r
t
h
e
s
i
m
p
l
e
s
u
b
t
a
s
k
s
.
Hidden Costs & Gotchas
- ⚠ {'title': 'GPT-4.1 Nano is now cheaper', 'detail': "GPT-4.1 Nano costs $0.10/$0.40 per million tokens versus Mini's $0.15/$0.60. That is 33% cheaper on input and output. For simple classification and routing, Nano may be the better budget pick — benchmark your specific task before choosing."}
- ⚠ {'title': 'Output tokens cost 4x input tokens', 'detail': 'GPT-4o Mini charges $0.15 per 1M input tokens but $0.60 per 1M output tokens. Generation-heavy tasks like summarization, content writing, or code generation cost more than you expect if you budget only on input. Always estimate your output-to-input ratio.'}
- ⚠ {'title': 'Fine-tuning training costs add up', 'detail': 'Fine-tuning training costs $0.30 per 1M tokens, and inference on fine-tuned models is 2x the base rate ($0.30/$1.20). A training run on 10M tokens costs $3.00 before you make a single inference call. Multiple training iterations multiply that cost.'}
- ⚠ {'title': '128K context does not mean 128K quality', 'detail': 'GPT-4o Mini accepts 128K tokens of context, but quality degrades noticeably past 32K tokens. Long-context tasks like document Q&A over large files may produce worse results than splitting into smaller chunks. Test with your actual data before relying on full context length.'}
- ⚠ {'title': 'Vision adds image token costs', 'detail': 'Sending images to GPT-4o Mini converts them to tokens based on resolution. A low-res image costs roughly $0.002 and a high-res image costs $0.004-0.006 depending on dimensions. If you process thousands of images, these costs add up fast even at Mini pricing.'}
- ⚠ {'title': 'Rate limits start low on free tier', 'detail': 'Free tier accounts get 500 RPM for GPT-4o Mini. Tier 1 ($5+ spent) stays at 500 RPM with 30K TPM. You need to reach Tier 2 ($50+ spent) for 5K RPM. Production applications often hit rate limits before they hit cost ceilings.'}
Which Plan Do You Need?
Prototyping and hobby projects
GPT-4o Mini at standard rates ($0.15/$0.60). Cheap enough to iterate without worrying about cost. 1,000 requests with 500 input and 200 output tokens each costs about $0.20.
Classification and routing at scale
GPT-4o Mini Batch at $0.075/$0.30. If latency does not matter, the Batch API cuts costs in half. Ideal for labeling datasets, content moderation, and intent classification on bulk data.
Production with quality requirements
Upgrade to GPT-4.1 ($2/$8) or GPT-4o ($2.50/$10). Mini struggles with complex reasoning, nuanced writing, and multi-step tool use. If accuracy matters more than cost, step up to a full-size model.
Budget production workloads
Compare GPT-4.1 Nano ($0.10/$0.40) versus GPT-4o Mini ($0.15/$0.60). Nano is 33% cheaper. Run both on your actual prompts and pick whichever scores higher — the cost difference is small but compounds at scale.
The Bottom Line
GPT-4o Mini at $0.15/$0.60 is excellent for simple tasks but now faces competition from GPT-4.1 Nano ($0.10/$0.40). For classification and routing, both are fine — pick based on benchmark performance for your specific task. For anything requiring reasoning or nuance, step up to GPT-4.1 or GPT-4o.
Related Resources
Frequently Asked Questions
How much does GPT-4o Mini cost?
GPT-4o Mini costs $0.15 per 1 million input tokens and $0.60 per 1 million output tokens. With prompt caching, input drops to $0.075/1M. Batch API pricing is $0.075 input and $0.30 output per million tokens.
Is GPT-4o Mini free?
No. GPT-4o Mini is not free through the API. New OpenAI accounts get $5 in credits which can be used for Mini. ChatGPT Free includes limited Mini access in the consumer product, but API usage always costs money.
GPT-4o Mini vs GPT-4.1 Nano — which is cheaper?
GPT-4.1 Nano is cheaper at $0.10/$0.40 per million tokens versus Mini's $0.15/$0.60. Nano is 33% less on both input and output. For new projects, benchmark both on your task — Nano is the better default unless Mini specifically outperforms.
Can I fine-tune GPT-4o Mini?
Yes. Fine-tuning GPT-4o Mini costs $0.30 per 1M training tokens. Inference on fine-tuned models costs $0.30 input and $1.20 output per million tokens — double the base rate. You need at least 10 training examples, though 50-100 examples typically produce better results.
What is GPT-4o Mini good for?
GPT-4o Mini excels at classification, entity extraction, content routing, simple Q&A, and any task where the output is short and structured. It matches GPT-4o closely on binary classification tasks while costing 94% less.
What is the GPT-4o Mini context window?
GPT-4o Mini supports 128K tokens of context (roughly 96K words). However, quality degrades past 32K tokens for most tasks. If you need reliable long-context performance, GPT-4.1 Nano offers 1M tokens.
Is GPT-4o Mini good for coding?
GPT-4o Mini handles simple code tasks like boilerplate generation, syntax fixes, and code explanation. It struggles with complex refactoring, multi-file changes, and architectural decisions. For serious coding, use GPT-4.1 or Claude Sonnet 4.6.
How does GPT-4o Mini compare to Claude Haiku?
Claude Haiku 4.5 costs $0.80/$4.00 per million tokens — significantly more expensive than GPT-4o Mini at $0.15/$0.60. Mini is roughly 5x cheaper on input and 7x cheaper on output. Haiku offers a 200K context window and is generally stronger on nuanced text tasks, but Mini wins on price for simple workloads.