OpenAI
·Released on Apr 16, 2025

O3 API Benchmarks, Pricing & Provider Data

Compare O3 with another model

Choose a model to open its comparison page.

Share on X
LLM

O3 benchmark, API pricing, and provider data cover 755 API providers, with prices starting at $0.034/request. O3 free API options are available from 1 provider. The page also shows measured API speed and first-token latency.

OpenAI O3 is a reasoning model in the O series, designed for complex reasoning, problem-solving, and analytical tasks.

Quality
#82of 111
50.0
LMSpeed score
Speed
#31of 80
134char/s
2.82 s
Cost
#142of 185
$0.034/ 1M · 8:1 in:out
$0.0055 in · $0.029 out

Category Performance

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

Coverage
6 / 8
Methodology
V3.0
Category PerformanceObserved-capability estimates with uncertainty reported separately from independent benchmark families.Agents49.1Coding53.1Reasoning54.2Knowledge47.6Math58.7Multilingual-Multimodal-Instruction following53
#1Math58.7Estimated3/4 Measured dimensions
80% interval: 49.467.9
foundational math66
math_500 · math_500
competition math61.7
aime · aime
proof frontier48.2
frontiermath · frontier_math_v2_tier4 / frontiermath · frontier_math_v2_tiers13
applied tool mathPrior only
#2Reasoning54.2RatedGlobal rank #223/4 Measured dimensions
80% interval: 45.562.8
abstract logicPrior only
scientific causal51.9
critpt · critpt / gpqa · gpqa / hle · hle
multistep constraints56.5
mmlu_pro · mmlu_pro
evidence verification54.1
lcr · lcr
#3Coding53.1Provisional1/4 Measured dimensions
80% interval: 39.267.0
code generation53.1
livecodebench · livecodebench / scicode · aa_sci_code
repository engineeringPrior only
debugging testingPrior only
tooling qualityPrior only
#4Instruction following53Provisional1/4 Measured dimensions
80% interval: 37.069.0
constraint followingPrior only
structured outputPrior only
novel instruction generalization53
ifbench · aa_if_bench
long multiturn instructionPrior only
#5Agents49.1Provisional1/4 Measured dimensions
80% interval: 33.165.1
planningPrior only
tool use49.1
tau · tau2_bench
environment executionPrior only
recovery reliabilityPrior only
#6Knowledge47.6Provisional1/4 Measured dimensions
80% interval: 31.663.6
broad knowledge47.6
aa_omniscience · aa_omniscience_index
professional knowledgePrior only
factualityPrior only
retrieval open bookPrior only
No data:MultilingualMultimodal

Specifications

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

INPUT
200Ktokens
240 pages of text
OUTPUT
100Ktokens
8K128K1M4M
200K

Features

Technical Details

Input
Output
Released
Apr 2025
Knowledge cutoff
2024-06-30
Documentation
Tokenizer
GPT
Architecture
text+image+file->text
Moderated
Yes
Supported parameters
include_reasoningmax_tokensreasoningresponse_formatseedstructured_outputstool_choicetools

Rankings

Excels at

It's decent at

Falls behind in

Detailed scores

Updated: Sep 5, 2026

Overall

undefined metric} other undefined metrics}}

Overall score50.0#82 / 111
Overall score
LMSpeed rank#82
Score50.0
UpdatedAug 20, 2026
Confidence1

Speed & latency

undefined metric} other undefined metrics}}

Output speed123.3 tok/s#31 / 80Time to first token5.17 s#64 / 80
Output speed
LMSpeed rank#31
Score123.3 tok/s
UpdatedSep 5, 2026
Confidence4
Time to first token
LMSpeed rank#64
Score5.17 s
UpdatedSep 5, 2026
Confidence4

Pricing

undefined metric} other undefined metrics}}

Input price$2.00/M#142 / 185Output price$8.00/M#137 / 185
Input price
LMSpeed rank#142
Score$2.00/M
UpdatedSep 5, 2026
Confidence4
Output price
LMSpeed rank#137
Score$8.00/M
UpdatedSep 5, 2026
Confidence4

Agents

V3.0

undefined metric} other undefined metrics}} · Provisional

Score49.180% interval33.1–65.11/4 Measured dimensionsΤ²-bench results80.7#54 / 82
Dimensions and evidence
Planning & decompositionPrior only
Tool use49.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
tau · tau2_bench · z -0.22 · q 1.00
Environment & long-horizon executionPrior only
Recovery & completion reliabilityPrior only
Τ²-bench resultsV3 evidence
LMSpeed rank#54
Score80.7
UpdatedAug 20, 2026
Confidence1

Coding

V3.0

undefined metric} other undefined metrics}} · Provisional

Score53.180% interval39.2–67.01/4 Measured dimensionsLiveCodeBench80.8%#16 / 115AA-SciCode41.0#67 / 113
Dimensions and evidence
Code generation53.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
livecodebench · livecodebench · z 0.99 · q 1.00
scicode · aa_sci_code · z -0.24 · q 1.00
Repository engineeringPrior only
Debugging & testingPrior only
Tool-assisted development & qualityPrior only
LiveCodeBenchV3 evidence
LMSpeed rank#16
Score80.8%
UpdatedSep 5, 2026
Confidence4
AA-SciCodeV3 evidence
LMSpeed rank#67
Score41.0
UpdatedAug 20, 2026
Confidence1

Reasoning

V3.0

undefined metric} other undefined metrics}} · Rated

Score54.2#2280% interval45.5–62.83/4 Measured dimensionsMMLU-Pro85.3%#27 / 129GPQA82.7%#82 / 217
Dimensions and evidence
Abstract logicPrior only
Scientific & causal reasoning51.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
critpt · critpt · z -0.41 · q 1.00
gpqa · gpqa · z 0.29 · q 1.00
hle · hle · z 0.27 · q 1.00
Multi-step constraints56.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmlu_pro · mmlu_pro · z 0.67 · q 1.00
Evidence integration & verification54.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
lcr · lcr · z 0.10 · q 1.00
MMLU-ProV3 evidence
LMSpeed rank#27
Score85.3%
UpdatedSep 5, 2026
Confidence4
GPQAV3 evidence
LMSpeed rank#82
Score82.7%
UpdatedSep 5, 2026
Confidence4
HLEV3 evidence
LMSpeed rank#84
Score20.1%
UpdatedSep 5, 2026
Confidence4
AA-LCRV3 evidence
LMSpeed rank#42
Score73.3
UpdatedAug 20, 2026
Confidence1
CritPtV3 evidence
LMSpeed rank#70
Score1.1
UpdatedAug 20, 2026
Confidence1

Knowledge

V3.0

undefined metric} other undefined metrics}} · Provisional

Score47.680% interval31.6–63.61/4 Measured dimensionsArtificial Analysis Intelligence Index31.1#75 / 117AA-GPQA Diamond82.7#71 / 113
Dimensions and evidence
Broad knowledge47.6
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aa_omniscience · aa_omniscience_index · z -0.40 · q 1.00
Professional knowledgePrior only
FactualityPrior only
Retrieval & open-book usePrior only
Artificial Analysis Intelligence IndexV3 evidence
LMSpeed rank#75
Score31.1
UpdatedAug 20, 2026
Confidence1
AA-GPQA Diamond
LMSpeed rank#71
Score82.7
UpdatedAug 20, 2026
Confidence1
AA-HLE
LMSpeed rank#72
Score20.1
UpdatedAug 20, 2026
Confidence1
AA-Omniscience Accuracy
LMSpeed rank#40
Score38.6
UpdatedAug 20, 2026
Confidence1
AA-Omniscience Hallucination Rate
LMSpeed rank#25
Score88.1
UpdatedAug 20, 2026
Confidence1

Math

V3.0

undefined metric} other undefined metrics}} · Estimated

Score58.780% interval49.4–67.93/4 Measured dimensionsAIME90.3%#3 / 68MATH-50099.2%#1 / 73
Dimensions and evidence
Foundational math66
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
math_500 · math_500 · z 2.17 · q 1.00
Competition math61.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aime · aime · z 1.54 · q 1.00
Advanced proofs48.2
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 -0.18 · q 1.00
Applied & tool-assisted mathPrior only
AIMEV3 evidence
LMSpeed rank#3
Score90.3%
UpdatedSep 5, 2026
Confidence4
MATH-500V3 evidence
LMSpeed rank#1
Score99.2%
UpdatedSep 5, 2026
Confidence4
Math score
LMSpeed rank#46
Score37.9
UpdatedAug 20, 2026
Confidence1
AA MATH-500
LMSpeed rank#1
Score99.2
UpdatedAug 20, 2026
Confidence1
FrontierMath v2 (Tiers 1-3)V3 evidence
LMSpeed rank#30
Score18.7
UpdatedAug 20, 2026
Confidence1
FrontierMath v2 (Tier 4)V3 evidence
LMSpeed rank#33
Score2.1
UpdatedAug 20, 2026
Confidence1

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}} · No data

80% interval30.8–69.20/4 Measured dimensionsAA-MMMU-Pro70.1#51 / 68Design Arena Website1047.0#71 / 78
Dimensions and evidence
Perception & OCRPrior only
Document & spatial understandingPrior only
Visual reasoningPrior only
Video & grounded actionPrior only
AA-MMMU-Pro
LMSpeed rank#51
Score70.1
UpdatedAug 20, 2026
Confidence1
Design Arena Website
LMSpeed rank#71
Score1047.0
UpdatedAug 20, 2026
Confidence1

Instruction following

V3.0

undefined metric} other undefined metrics}} · Provisional

Score5380% interval37.0–69.01/4 Measured dimensionsAA-IFBench71.4#31 / 84
Dimensions and evidence
Constraint followingPrior only
Structured outputPrior only
Novel-instruction generalization53
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifbench · aa_if_bench · z 0.12 · q 1.00
Multi-turn & long instructionsPrior only
AA-IFBenchV3 evidence
LMSpeed rank#31
Score71.4
UpdatedAug 20, 2026
Confidence1

OpenRouter endpoints

1 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
$2/M$8/M100%undefined tokens / undefined tokens

Pricing Comparison

Compare O3 API pricing across 754 providers. Prices range from $0.034/request to $300.00/M. KFCV50 offers the lowest rate at $0.034/request. 1 provider offers free API credits or a free tier.

ProviderHealthModel VariantGroupInput ($/M)Output ($/M)Speed (t/s)First tokenAudit
L1
100%
o3
default
-93%$0.137/M
Cache read$0.034/M
-93%$0.548/M
L1
100%
o3-2025-04-16
default
-93%$0.137/M
Cache read$0.034/M
-93%$0.548/M
L1
100%
o3
default
$2.00/M
$8.00/M
L1
100%
o3-2025-04-16
default
$2.00/M
$8.00/M
L1
100%
o3
default
$0.068/request
-
L1
100%
o3-all
default
$0.068/request
-
L1
100%
o3-2025-04-16
default
-86%$0.274/M
-86%$1.10/M
L1
100%
o3-medium
az_stable_high_22
-86%$0.274/M
-86%$1.10/M
L1
100%
o3-2025-04-16-high
az_stable_high_22
-86%$0.274/M
-86%$1.10/M
L1
100%
o3-2025-04-16-low
az_stable_high_22
-86%$0.274/M
-86%$1.10/M
L1
100%
o3-2025-04-16-medium
az_stable_high_22
-86%$0.274/M
-86%$1.10/M
L1
100%
o3-high
az_stable_high_22
-86%$0.274/M
-86%$1.10/M
L1
100%
o3-low
az_stable_high_22
-86%$0.274/M
-86%$1.10/M
L1
100%
o3-2025-04-16
default
-23%$1.53/M
-23%$6.14/M
L1
100%
o3
default
-23%$1.53/M
-23%$6.14/M
L1
100%
o3-2025-04-16
原价分组
Free
Free
L1
100%
o3-all
default
$0.600/request
-
L1
100%
o3
default
$0.600/request
-
L1
100%
o3-2025-04-16
default
-96%$0.082/M
Cache read$0.021/M
-96%$0.329/M
L1
100%
o3
default
-96%$0.082/M
Cache read$0.021/M
-96%$0.329/M
Showing 20 model IDs of 132.

Alternatives & Similar Models

Frequently Asked Questions

What benchmark data does O3 include?
LMSpeed shows O3 benchmark context, API price, output speed, first-token latency, and provider data across 755 providers when those signals are available.
What is the O3 API price?
O3 has pricing from 755 providers, ranging from $0.034/request to $300.00/M. KFCV50 has the lowest listed price.
What does the O3 API pricing table include?
The O3 API pricing table compares 755 providers by input price, output price, free tier status, speed, first-token latency, and recent health data when available.
Which provider has the cheapest O3 API pricing?
KFCV50 currently has the lowest listed O3 price at $0.034/request across 755 providers.
Can I compare O3 API price and speed together?
Yes. LMSpeed shows O3 API price, output speed, first-token latency, and provider health on the same page so you can compare cost and performance together.
Is O3 API free?
Yes, O3 free API options are available through 1 providerundefined other undefined} on LMSpeed, including 兔子API. These providers offer free API credits or a free tier with no per-token charges.
Where can I get O3 free API access?
LMSpeed currently lists 1 free API providerundefined other undefined} for O3: 兔子API. Check each provider row before using it because free tier limits can change.

Also known as

o3o3-2025-04-16o3-2025-04-16-higho3-2025-04-16-lowo3-2025-04-16-medium

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