OpenAI
·Released on Mar 17, 2026

GPT-5.4 Mini API Benchmarks, Pricing & Provider Data

Compare GPT-5.4 Mini with another model

Choose a model to open its comparison page.

Share on X
LLM

GPT-5.4 Mini benchmark, API pricing, and provider data cover undefined API provider} other undefined API providers}}, with prices starting at $0.00003/request. GPT-5.4 Mini free API options are available from undefined provider} other undefined providers}}. The page also shows measured API speed and first-token latency.

OpenAI GPT-5.4 Mini is a compact language model in the GPT-5 series, optimized for quick responses and high throughput.

Quality
#48of 112
60.0
LMSpeed score
Speed
134char/s
3.54 s
Cost
#107of 186
$0.00003/ 1M · 8:1 in:out
$0.0000048 in · $0.0000252 out

Category Performance

Observed-capability estimates with uncertainty reported separately from independent benchmark families.

Coverage
7 / 8
Methodology
V3.0
Category PerformanceObserved-capability estimates with uncertainty reported separately from independent benchmark families.Agents50.1Coding56.8Reasoning55.4Knowledge45.4Math49.3Multilingual-Multimodal47.1Instruction following53.8
#1Coding56.8Estimated2/4 Measured dimensions
80% interval: 44.968.6
code generation57.8
scicode · scicode
repository engineering55.8
vibecode · vibe_code_bench
debugging testingPrior only
tooling qualityPrior only
#2Reasoning55.4Estimated2/4 Measured dimensions
80% interval: 44.666.2
abstract logicPrior only
scientific causal56.4
critpt · critpt / gpqa · gpqa / hle · hle / hle · hle_no_tools
multistep constraintsPrior only
evidence verification54.4
lcr · lcr
#3Instruction following53.8Provisional1/4 Measured dimensions
80% interval: 37.869.8
constraint followingPrior only
structured outputPrior only
novel instruction generalization53.8
ifbench · aa_if_bench
long multiturn instructionPrior only
#4Agents50.1RatedGlobal rank #393/4 Measured dimensions
80% interval: 42.058.2
planningPrior only
tool use47.6
mcp_atlas · mcp_atlas / tau · tau2_bench / toolathlon · toolathlon
environment execution49.3
gdpval_aa · benchlm_agentic_gdpval_aa / osworld · os_world_verified / terminalbench · benchlm_agentic_terminal_bench2
recovery reliability53.4
apex_agents · apex_agents_aa
#5Math49.3Provisional1/4 Measured dimensions
80% interval: 33.365.4
foundational mathPrior only
competition mathPrior only
proof frontier49.3
frontiermath · frontier_math_v2_tier4 / frontiermath · frontier_math_v2_tiers13
applied tool mathPrior only
#6Multimodal47.1Provisional1/4 Measured dimensions
80% interval: 31.063.3
perception ocrPrior only
document spatialPrior only
visual reasoning47.1
mmmu_pro · mmmu_pro
video actionPrior only
#7Knowledge45.4Provisional1/4 Measured dimensions
80% interval: 31.459.4
broad knowledge45.4
aa_omniscience · aa_omniscience_index / benchlm_category_knowledge · benchlm_category_knowledge
professional knowledgePrior only
factualityPrior only
retrieval open bookPrior only
No data:Multilingual

Specifications

Input and output token limits for this model, plus how it ranks on long-context understanding.

INPUT
400Ktokens
480 pages of text
OUTPUT
128Ktokens
8K128K1M4M
400K

Features

Technical Details

Input
Output
Released
Mar 2026
Knowledge cutoff
2025-08-31
Documentation
Tokenizer
GPT
Architecture
text+image+file->text
Moderated
Yes
Supported parameters
include_reasoningmax_completion_tokensmax_tokensreasoningreasoning_effortresponse_formatseedstructured_outputstool_choicetools

Rankings

Excels at

It's decent at

Falls behind in

Detailed scores

Updated: Sep 13, 2026

Overall

undefined metric} other undefined metrics}}

Overall score60.0#48 / 112
Overall score
LMSpeed rank#48
Score60.0
UpdatedSep 13, 2026
Confidence4

Pricing

undefined metric} other undefined metrics}}

Input price$0.750/M#107 / 186Output price$4.50/M#125 / 186
Input price
LMSpeed rank#107
Score$0.750/M
UpdatedSep 13, 2026
Confidence4
Output price
LMSpeed rank#125
Score$4.50/M
UpdatedSep 13, 2026
Confidence4

Agents

V3.0

undefined metric} other undefined metrics}} · Rated

Score50.1#3980% interval42.0–58.23/4 Measured dimensionsAgentic score55.6#49 / 77Terminal-Bench 2.060.0#31 / 53
Dimensions and evidence
Planning & decompositionPrior only
Tool use47.6
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mcp_atlas · mcp_atlas · z -1.30 · q 1.00
tau · tau2_bench · z 0.46 · q 1.00
toolathlon · toolathlon · z -0.52 · q 1.00
Environment & long-horizon execution49.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
gdpval_aa · benchlm_agentic_gdpval_aa · z -0.18 · q 1.00
osworld · os_world_verified · z -0.09 · q 1.00
terminalbench · benchlm_agentic_terminal_bench2 · z -0.24 · q 1.00
Recovery & completion reliability53.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
apex_agents · apex_agents_aa · z 0.21 · q 1.00
Agentic score
LMSpeed rank#49
Score55.6
UpdatedSep 13, 2026
Confidence4
Terminal-Bench 2.0V3 evidence
LMSpeed rank#31
Score60.0
UpdatedSep 13, 2026
Confidence4
OSWorld-VerifiedV3 evidence
LMSpeed rank#16
Score72.1
UpdatedSep 13, 2026
Confidence4
MCP AtlasV3 evidence
LMSpeed rank#24
Score57.7
UpdatedSep 13, 2026
Confidence4
ToolathlonV3 evidence
LMSpeed rank#18
Score42.9
UpdatedSep 13, 2026
Confidence4
Τ²-bench resultsV3 evidence
LMSpeed rank#28
Score93.4
UpdatedSep 13, 2026
Confidence4
AA Agentic Index
LMSpeed rank#50
Score19.6
UpdatedSep 13, 2026
Confidence4
APEX-Agents-AAV3 evidence
LMSpeed rank#11
Score28.2
UpdatedSep 13, 2026
Confidence4
GDPval-AA
LMSpeed rank#40
Score29.7
UpdatedSep 13, 2026
Confidence4
GDPval-AAV3 evidence
LMSpeed rank#42
Score1095.0
UpdatedSep 13, 2026
Confidence4
Terminal-Bench 2.1 (Vals)
LMSpeed rank#29
Score54.7
UpdatedSep 13, 2026
Confidence4

Coding

V3.0

undefined metric} other undefined metrics}} · Estimated

Score56.880% interval44.9–68.62/4 Measured dimensionsSciCode52.1%#25 / 89Coding score56.0#36 / 87
Dimensions and evidence
Code generation57.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
scicode · scicode · z 0.87 · q 1.00
Repository engineering55.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
vibecode · vibe_code_bench · z 0.79 · q 1.00
Debugging & testingPrior only
Tool-assisted development & qualityPrior only
SciCodeV3 evidence
LMSpeed rank#25
Score52.1%
UpdatedSep 13, 2026
Confidence4
Coding score
LMSpeed rank#36
Score56.0
UpdatedSep 13, 2026
Confidence4
Vibe Code BenchV3 evidence
LMSpeed rank#10
Score48.0
UpdatedSep 13, 2026
Confidence4
AA Coding Index
LMSpeed rank#36
Score56.1
UpdatedSep 13, 2026
Confidence4
AA-SciCodeV3 evidence
LMSpeed rank#28
Score52.1
UpdatedSep 13, 2026
Confidence4
FrontierCode 1.1 Main
LMSpeed rank#9
Score27.0
UpdatedSep 13, 2026
Confidence4
LiveCodeBench (Vals)
LMSpeed rank#31
Score81.5
UpdatedSep 13, 2026
Confidence4
SWE-bench (Vals)
LMSpeed rank#33
Score73.0
UpdatedSep 13, 2026
Confidence4
EEBench
LMSpeed rank#16
Score18.3
UpdatedAug 20, 2026
Confidence3

Reasoning

V3.0

undefined metric} other undefined metrics}} · Estimated

Score55.480% interval44.6–66.22/4 Measured dimensionsGPQA82.3%#87 / 218HLE18.6%#91 / 216
Dimensions and evidence
Abstract logicPrior only
Scientific & causal reasoning56.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
critpt · critpt · z 0.48 · q 1.00
gpqa · gpqa · z 0.65 · q 1.00
hle · hle · z 0.52 · q 1.00
hle · hle_no_tools · z -1.74 · q 1.00
Multi-step constraintsPrior only
Evidence integration & verification54.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
lcr · lcr · z 0.23 · q 1.00
GPQAV3 evidence
LMSpeed rank#87
Score82.3%
UpdatedSep 13, 2026
Confidence4
HLEV3 evidence
LMSpeed rank#91
Score18.6%
UpdatedSep 13, 2026
Confidence4
AA-LCRV3 evidence
LMSpeed rank#43
Score77.0
UpdatedSep 13, 2026
Confidence4
CritPtV3 evidence
LMSpeed rank#33
Score10.0
UpdatedSep 13, 2026
Confidence4
Reasoning score
LMSpeed rank#29
Score73.9
UpdatedSep 13, 2026
Confidence4

Knowledge

V3.0

undefined metric} other undefined metrics}} · Provisional

Score45.480% interval31.4–59.41/4 Measured dimensionsKnowledge score57.0#62 / 83HLE w/o tools28.2#21 / 22
Dimensions and evidence
Broad knowledge45.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aa_omniscience · aa_omniscience_index · z -0.52 · q 1.00
benchlm_category_knowledge · benchlm_category_knowledge · z -0.76 · q 1.00
Professional knowledgePrior only
FactualityPrior only
Retrieval & open-book usePrior only
Knowledge scoreV3 evidence
LMSpeed rank#62
Score57.0
UpdatedSep 13, 2026
Confidence4
HLE w/o tools
LMSpeed rank#21
Score28.2
UpdatedSep 13, 2026
Confidence4
Artificial Analysis Intelligence IndexV3 evidence
LMSpeed rank#69
Score24.6
UpdatedSep 13, 2026
Confidence4
AA-GPQA Diamond
LMSpeed rank#46
Score87.5
UpdatedSep 13, 2026
Confidence4
AA-HLE
LMSpeed rank#54
Score28.1
UpdatedSep 13, 2026
Confidence4
AA-Omniscience Accuracy
LMSpeed rank#44
Score37.5
UpdatedSep 13, 2026
Confidence4
AA-Omniscience Hallucination Rate
LMSpeed rank#18
Score90.2
UpdatedSep 13, 2026
Confidence4
GPQA Diamond (Vals)
LMSpeed rank#35
Score83.1
UpdatedSep 13, 2026
Confidence4
MMLU-Pro (Vals)
LMSpeed rank#32
Score84.6
UpdatedSep 13, 2026
Confidence4

Math

V3.0

undefined metric} other undefined metrics}} · Provisional

Score49.380% interval33.3–65.41/4 Measured dimensionsMath score44.5#39 / 63FrontierMath v2 (Tiers 1-3)28.3#20 / 47
Dimensions and evidence
Foundational mathPrior only
Competition mathPrior only
Advanced proofs49.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
frontiermath · frontier_math_v2_tier4 · z -0.60 · q 1.00
frontiermath · frontier_math_v2_tiers13 · z 0.13 · q 1.00
Applied & tool-assisted mathPrior only
Math score
LMSpeed rank#39
Score44.5
UpdatedSep 13, 2026
Confidence4
FrontierMath v2 (Tiers 1-3)V3 evidence
LMSpeed rank#20
Score28.3
UpdatedSep 13, 2026
Confidence4
FrontierMath v2 (Tier 4)V3 evidence
LMSpeed rank#36
Score2.1
UpdatedSep 13, 2026
Confidence4

Multilingual

V3.0

undefined metric} other undefined metrics}} · No data

No data80% interval30.8–69.20/4 Measured dimensions
Dimensions and evidence
Cross-language understandingPrior only
Multilingual generationPrior only
Reasoning transferPrior only
Low-resource robustnessPrior only

Multimodal

V3.0

undefined metric} other undefined metrics}} · Provisional

Score47.180% interval31.0–63.31/4 Measured dimensionsMultimodal Grounded score57.2#41 / 56MMMU-Pro76.6#22 / 31
Dimensions and evidence
Perception & OCRPrior only
Document & spatial understandingPrior only
Visual reasoning47.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmmu_pro · mmmu_pro · z -0.45 · q 1.00
Video & grounded actionPrior only
Multimodal Grounded score
LMSpeed rank#41
Score57.2
UpdatedSep 13, 2026
Confidence4
MMMU-ProV3 evidence
LMSpeed rank#22
Score76.6
UpdatedSep 13, 2026
Confidence4
MMMU-Pro w/ Python
LMSpeed rank#8
Score78.0
UpdatedSep 13, 2026
Confidence4
AA-MMMU-Pro
LMSpeed rank#42
Score73.3
UpdatedSep 13, 2026
Confidence4

Instruction following

V3.0

undefined metric} other undefined metrics}} · Provisional

Score53.880% interval37.8–69.81/4 Measured dimensionsAA-IFBench73.3#23 / 84Instruction Following score89.8#17 / 52
Dimensions and evidence
Constraint followingPrior only
Structured outputPrior only
Novel-instruction generalization53.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifbench · aa_if_bench · z 0.21 · q 1.00
Multi-turn & long instructionsPrior only
AA-IFBenchV3 evidence
LMSpeed rank#23
Score73.3
UpdatedSep 13, 2026
Confidence4
Instruction Following score
LMSpeed rank#17
Score89.8
UpdatedSep 13, 2026
Confidence4

OpenRouter endpoints

5 endpoints

Third-party OpenRouter endpoint data, shown separately from LMSpeed measurements. Some 30-minute live performance fields only appear after syncing with an OpenRouter API key.

Provider endpointInputOutput1d uptime30m latency30m throughputContext / output
OpenAI
openai/flex
$0.375/M$2.25/M100%undefined tokens / undefined tokens
Azure
azure/us
$0.825/M$4.95/M100%undefined tokens / undefined tokens
OpenAI
openai/fast
$1.50/M$9/M100%undefined tokens / undefined tokens
Azure
azure
$0.750/M$4.50/M100.0%undefined tokens / undefined tokens
OpenAI
openai
$0.750/M$4.50/M99.9%undefined tokens / undefined tokens

Pricing Comparison

Compare GPT-5.4 Mini API pricing across 452 providers. Prices range from $0.00003/request to $300.00/M. CM-API 公益站 offers the lowest rate at $0.00003/request. 9 providers offer free API credits or a free tier.

ProviderHealthModel VariantGroupInput ($/M)Output ($/M)Speed (t/s)First tokenAudit
L1
0%
gpt-5.4-mini
default
$0.750/M
Cache read$0.075/M
$4.50/M
245.0 t/s
2.62 s
L1
100%
gpt-5.4-mini
default
-93%$0.051/M
Cache read$0.0051/M
-93%$0.308/M
200.7 t/s
1.44 s
L1
100%
gpt-5.4-mini-2026-03-17
default
-93%$0.051/M
-93%$0.308/M
L1
96%
gpt-5.4-mini
default
$0.750/M
Cache read$0.075/M
$4.50/M
139.4 t/s
2.84 s
L1
99%
gpt-5.4-mini
default
$0.945/M
Cache read$0.945/MCache write$0.945/MCache write 1h$1.51/M
-79%$0.945/M
41.2 t/s
3.66 s
L1
100%
gpt-5.4-mini
default
$0.750/M
$4.50/M
L1
100%
gpt-5.4-mini-xhigh
default
$0.750/M
$4.50/M
L1
100%
gpt-5.4-mini-medium
default
$0.750/M
$4.50/M
L1
100%
gpt-5.4-mini-low
default
$0.750/M
$4.50/M
L1
100%
gpt-5.4-mini-high
default
$0.750/M
$4.50/M
L1
100%
gpt-5.4-mini-2026-03-17
default
$0.750/M
$4.50/M
L1
100%
gpt-5.4-mini
GPT · Best cash equivalent (conditional promotion)
-95%$0.037/M
Cache read$0.0037/M
-95%$0.225/M
L1
100%
gpt-5.4-mini
azure-officially
-29%$0.536/M
Cache read$0.054/M
-29%$3.21/M
L1
100%
gpt-5.4-mini
default
-93%$0.051/M
Cache read$0.0051/M
-93%$0.308/M
L1
100%
gpt-5.4-mini-2026-03-17
default
-93%$0.051/M
-93%$0.308/M
L1
100%
gpt-5.4-mini
default
-86%$0.103/M
Cache read$0.010/M
-86%$0.616/M
L1
100%
gpt-5.4-mini-2026-03-17
default
-86%$0.103/M
Cache read$0.010/M
-86%$0.616/M
L1
100%
gpt-5.4-mini
default
-23%$0.575/M
-23%$3.45/M
L1
100%
gpt-5.4-mini
openai-offical
$0.750/M
Cache read$0.075/M
$4.50/M
L1
100%
gpt-5.4-mini
default
$0.015/request
-
Showing 20 model IDs of 142.

Alternatives & Similar Models

Frequently Asked Questions

What benchmark data does GPT-5.4 Mini include?
LMSpeed shows GPT-5.4 Mini benchmark context, API price, output speed, first-token latency, and provider data across 461 providers when those signals are available.
What is the GPT-5.4 Mini API price?
GPT-5.4 Mini has pricing from undefined provider} other undefined providers}}, ranging from $0.00003/request to $300.00/M. CM-API 公益站 has the lowest listed price.
What does the GPT-5.4 Mini API pricing table include?
The GPT-5.4 Mini API pricing table compares 461 providers by input price, output price, free tier status, speed, first-token latency, and recent health data when available.
Which provider has the cheapest GPT-5.4 Mini API pricing?
CM-API 公益站 currently has the lowest listed GPT-5.4 Mini price at $0.00003/request across undefined provider} other undefined providers}}.
Can I compare GPT-5.4 Mini API price and speed together?
Yes. LMSpeed shows GPT-5.4 Mini API price, output speed, first-token latency, and provider health on the same page so you can compare cost and performance together.
Is GPT-5.4 Mini API free?
Yes, GPT-5.4 Mini free API options are available through 9 providerundefined other undefined} on LMSpeed, including 兔子API, 兔子API, 兔子API, 兔子API, 兔子API. These providers offer free API credits or a free tier with no per-token charges.
Where can I get GPT-5.4 Mini free API access?
LMSpeed currently lists 9 free API providerundefined other undefined} for GPT-5.4 Mini: 兔子API, 兔子API, 兔子API, 兔子API, 兔子API. Check each provider row before using it because free tier limits can change.

Also known as

11/gpt-5.4-mini15/gpt-5.4-mini64/gpt-5.4-miniCPA/gpt-5.4-miniGPT-5.4 Mini

Rankings are based on community-submitted tests and periodic health probes. Advisory only, not official data.Standard benchmark data may include BenchLM and other public sources.

Build with LMSpeed Data

Free

Free public API for LLM pricing, benchmarks & provider data

  • Real-time pricing & availability
  • Speed & latency benchmarks
  • 300+ models, 600+ providers
  • 1,000 requests/day
View API Documentation