OpenAI
·Released on Mar 5, 2026GPT-5.4 API Benchmarks, Pricing & Provider Data
Compare GPT-5.4 with another model
Choose a model to open its comparison page.
LLM
GPT-5.4 benchmark, API pricing, and provider data cover undefined API provider} other undefined API providers}}, with prices starting at $0.0054/M. GPT-5.4 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 extends the GPT-5 family with stronger instruction following, deeper tool use, and improved performance on coding, math, and long-document analysis.
Quality
#41of 112
63.0
LMSpeed score
Speed
49char/s
4.45 s
Cost
#154of 186
$0.0054/ 1M · 8:1 in:out
$0.0009 in · $0.0045 out
Category Performance
Observed-capability estimates with uncertainty reported separately from independent benchmark families.
- Coverage
- 7 / 8
- Methodology
- V3.0
#1Coding62.2Estimated2/4 Measured dimensions
80% interval: 51.9–72.5
code generation65.5
livecodebench · live_code_bench_pro / scicode · aa_sci_code
repository engineering59
react_native_bench · react_native_evals / swe_pro · swe_pro / vibecode · vibe_code_bench
debugging testingPrior only
tooling qualityPrior only
#2Knowledge58.5Provisional1/4 Measured dimensions
80% interval: 43.5–73.6
broad knowledgePrior only
professional knowledge58.5
healthbench · health_bench_hard / medxpert · med_xpert_qa_text
factualityPrior only
retrieval open bookPrior only
#3Math56.8Provisional1/4 Measured dimensions
80% interval: 40.7–72.9
foundational mathPrior only
competition mathPrior only
proof frontier56.8
frontiermath · frontier_math_v2_tier4 / frontiermath · frontier_math_v2_tiers13
applied tool mathPrior only
#4Agents56.3RatedGlobal rank #194/4 Measured dimensions
80% interval: 51.2–61.4
planning58.3
deep_planning · gert_labs
tool use59.8
mcp_atlas · mcp_atlas / tau · tau2_bench / toolathlon · toolathlon
environment execution54.5
browsecomp · browse_comp / deep_search_qa · deep_search_qa / gdpval_aa · benchlm_agentic_gdpval_aa / osworld · os_world_verified / terminalbench · benchlm_agentic_terminal_bench2
recovery reliability52.6
apex_agents · apex_agents_aa / claw_eval · claw_eval / cyber_gym · cyber_gym / exploit_gym · exploit_gym / jobbench · job_bench / researchclaw · research_claw_bench
#5Reasoning55.7RatedGlobal rank #163/4 Measured dimensions
80% interval: 46.9–64.4
abstract logic50.7
arc_agi · arc_agi2
scientific causal59.1
critpt · critpt / gpqa · gpqa / hle · hle / hle · hle_no_tools
multistep constraintsPrior only
evidence verification57.2
lcr · lcr
#6Instruction following54Provisional1/4 Measured dimensions
80% interval: 38.0–70.0
constraint followingPrior only
structured outputPrior only
novel instruction generalization54
ifbench · aa_if_bench
long multiturn instructionPrior only
#7Multimodal52.5RatedGlobal rank #44/4 Measured dimensions
80% interval: 45.9–59.2
perception ocr46.7
simplevqa · simple_vqa
document spatial50.1
charxiv · charxiv
visual reasoning56.9
erqa · erqa / medxpert · med_xpert_qa_mm / mmmu_pro · mmmu_pro
video action56.5
screenspot · screen_spot_pro
No data:Multilingual
Specifications
Input and output token limits for this model, plus how it ranks on long-context understanding.
INPUT
1.1Mtokens
≈ 1.3K pages of text
OUTPUT
128Ktokens
8K128K1M4M
1.1M
Features
Rankings
Excels at
It's decent at
Falls behind in
Detailed scores
Updated: Sep 13, 2026Overall
undefined metric} other undefined metrics}}
Overall score63.0#41 / 112
Overall
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Pricing
undefined metric} other undefined metrics}}
Input price$2.50/M#154 / 186Output price$15.00/M#159 / 186
Pricing
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Score56.3#1980% interval51.2–61.44/4 Measured dimensionsAgentic score72.0#29 / 77Terminal-Bench 2.075.1#13 / 53
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Planning & decomposition58.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
deep_planning · gert_labs · z 1.12 · q 1.00
Tool use59.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mcp_atlas · mcp_atlas · z -0.26 · q 1.00
tau · tau2_bench · z 1.50 · q 1.00
toolathlon · toolathlon · z 0.67 · q 1.00
Environment & long-horizon execution54.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
browsecomp · browse_comp · z 0.11 · q 1.00
deep_search_qa · deep_search_qa · z -0.52 · q 1.00
gdpval_aa · benchlm_agentic_gdpval_aa · z 0.36 · q 1.00
osworld · os_world_verified · z 0.18 · q 1.00
terminalbench · benchlm_agentic_terminal_bench2 · z 0.79 · q 1.00
Recovery & completion reliability52.6
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
apex_agents · apex_agents_aa · z 0.51 · q 1.00
claw_eval · claw_eval · z 0.11 · q 1.00
cyber_gym · cyber_gym · z 0.18 · q 1.00
exploit_gym · exploit_gym · z -0.84 · q 0.58
jobbench · job_bench · z 0.27 · q 1.00
researchclaw · research_claw_bench · z -0.89 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Coding
V3.0undefined metric} other undefined metrics}} · Estimated
Score62.280% interval51.9–72.52/4 Measured dimensionsCoding score42.8#69 / 87LiveCodeBench Pro87.5#1 / 4
Coding
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Code generation65.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
livecodebench · live_code_bench_pro · z 1.29 · q 0.50
scicode · aa_sci_code · z 1.31 · q 1.00
Repository engineering59
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
react_native_bench · react_native_evals · z 0.98 · q 1.00
swe_pro · swe_pro · z 0.11 · q 1.00
vibecode · vibe_code_bench · z 1.43 · q 1.00
Debugging & testingPrior only
Tool-assisted development & qualityPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Reasoning
V3.0undefined metric} other undefined metrics}} · Rated
Score55.7#1680% interval46.9–64.43/4 Measured dimensionsGPQA74.8%#122 / 218HLE11.3%#116 / 216
Reasoning
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Abstract logic50.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
arc_agi · arc_agi2 · z 0.00 · q 1.00
Scientific & causal reasoning59.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
critpt · critpt · z 0.87 · q 1.00
gpqa · gpqa · z 0.61 · q 1.00
hle · hle · z 0.60 · q 1.00
hle · hle_no_tools · z -0.33 · q 1.00
Multi-step constraintsPrior only
Evidence integration & verification57.2
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
lcr · lcr · z 0.62 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Score58.580% interval43.5–73.61/4 Measured dimensionsKnowledge score74.3#30 / 83HLE w/o tools39.8#15 / 22
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Broad knowledgePrior only
Professional knowledge58.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
healthbench · health_bench_hard · z 0.66 · q 1.00
medxpert · med_xpert_qa_text · z 0.45 · q 0.50
FactualityPrior only
Retrieval & open-book usePrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Artificial Analysis Intelligence Index
LMSpeed rank#29
Score39.0
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence4
AA-Omniscience Hallucination Rate
LMSpeed rank#13
Score91.7
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence4
Math
V3.0undefined metric} other undefined metrics}} · Provisional
Score56.880% interval40.7–72.91/4 Measured dimensionsMath score64.5#24 / 63FrontierMath v2 (Tiers 1-3)47.6#7 / 47
Math
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Foundational mathPrior only
Competition mathPrior only
Advanced proofs56.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
frontiermath · frontier_math_v2_tier4 · z 0.90 · q 1.00
frontiermath · frontier_math_v2_tiers13 · z 0.61 · q 1.00
Applied & tool-assisted mathPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
FrontierMath v2 (Tiers 1-3)V3 evidence
LMSpeed rank#7
Score47.6
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence4
FrontierMath v2 (Tier 4)V3 evidence
LMSpeed rank#9
Score27.1
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence4
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
No data80% interval30.8–69.20/4 Measured dimensions
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
Dimensions and evidence
Cross-language understandingPrior only
Multilingual generationPrior only
Reasoning transferPrior only
Low-resource robustnessPrior only
Multimodal
V3.0undefined metric} other undefined metrics}} · Rated
Score52.5#480% interval45.9–59.24/4 Measured dimensionsMultimodal Grounded score69.3#28 / 56MMMU-Pro81.2#7 / 31
Multimodal
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Perception & OCR46.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
simplevqa · simple_vqa · z -0.43 · q 1.00
Document & spatial understanding50.1
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
charxiv · charxiv · z 0.17 · q 1.00
Visual reasoning56.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
erqa · erqa · z -0.01 · q 1.00
medxpert · med_xpert_qa_mm · z 0.33 · q 0.75
mmmu_pro · mmmu_pro · z 0.69 · q 1.00
Video & grounded action56.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
screenspot · screen_spot_pro · z 0.53 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Instruction following
V3.0undefined metric} other undefined metrics}} · Provisional
Score5480% interval38.0–70.01/4 Measured dimensionsAA-IFBench73.9#21 / 84Instruction Following score90.6#16 / 52
Instruction following
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Constraint followingPrior only
Structured outputPrior only
Novel-instruction generalization54
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifbench · aa_if_bench · z 0.24 · q 1.00
Multi-turn & long instructionsPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
OpenRouter endpoints
7 endpointsThird-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 endpoint | Input | Output | 1d uptime | 30m latency | 30m throughput | Context / output |
|---|---|---|---|---|---|---|
OpenAI openai/fast | $5/M | $30/M | 100% | — | — | undefined tokens / undefined tokens |
Azure azure | $2.50/M | $15/M | 100.0% | — | — | undefined tokens / undefined tokens |
OpenAI openai/flex | $1.25/M | $7.50/M | 99.0% | — | — | undefined tokens / undefined tokens |
OpenAI openai | $2.50/M | $15/M | 98.3% | — | — | undefined tokens / undefined tokens |
Azure azure/eu | $2.75/M | $16.50/M | — | — | — | undefined tokens / undefined tokens |
Azure azure/us | $2.75/M | $16.50/M | — | — | — | undefined tokens / undefined tokens |
Amazon Bedrock amazon-bedrock/us-east-1 | $2.75/M | $16.50/M | — | — | — | undefined tokens / undefined tokens |
Pricing Comparison
Compare GPT-5.4 API pricing across 569 providers. Prices range from $0.0054/M to $7425.00/M. 熊猫 API offers the lowest rate at $0.0054/M. 17 providers offer free API credits or a free tier.
Showing 20 model IDs of 214.
Alternatives & Similar Models
GPT-5.3 Codex
gpt-5-3-codex
OpenAI GPT-5.3 Codex is a code-specialized variant in the GPT-5 series, optimized for code generation, debugging, and software development tasks.
317 shared providers
GPT-5.2
gpt-5-2
OpenAI GPT-5.2 is a GPT-5 series model emphasizing advanced reasoning, multimodal understanding, and high-quality outputs for complex enterprise workloads.
305 shared providers
GPT-5.4 Mini
gpt-5-4-mini
OpenAI GPT-5.4 Mini is a compact language model in the GPT-5 series, optimized for quick responses and high throughput.
300 shared providers
Claude Opus 4.6
claude-opus-4-6
Anthropic Claude Opus 4.6 is the most capable Claude Opus tier, optimized for complex analysis, long-horizon coding, and high-stakes enterprise reasoning workloads.
270 shared providers
Claude Sonnet 4.6
claude-sonnet-4-6
Anthropic Claude Sonnet 4.6 extends the Sonnet line with improved tool use, coding reliability, and long-context performance for everyday production workloads.
264 shared providers
Claude Haiku 4.5
claude-haiku-4-5
Anthropic Claude Haiku 4.5 delivers fast, low-cost responses while retaining solid instruction following for chat, classification, and lightweight coding.
244 shared providers
Frequently Asked Questions
- What benchmark data does GPT-5.4 include?
- LMSpeed shows GPT-5.4 benchmark context, API price, output speed, first-token latency, and provider data across 586 providers when those signals are available.
- What is the GPT-5.4 API price?
- GPT-5.4 has pricing from undefined provider} other undefined providers}}, ranging from $0.0054/M to $7425.00/M. 熊猫 API has the lowest listed price.
- What does the GPT-5.4 API pricing table include?
- The GPT-5.4 API pricing table compares 586 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 API pricing?
- 熊猫 API currently has the lowest listed GPT-5.4 price at $0.0054/M across undefined provider} other undefined providers}}.
- Can I compare GPT-5.4 API price and speed together?
- Yes. LMSpeed shows GPT-5.4 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 API free?
- Yes, GPT-5.4 free API options are available through 17 providerundefined other undefined} on LMSpeed, including 兔子API, 兔子API, 兔子API, 兔子API, 180txt API. These providers offer free API credits or a free tier with no per-token charges.
- Where can I get GPT-5.4 free API access?
- LMSpeed currently lists 17 free API providerundefined other undefined} for GPT-5.4: 兔子API, 兔子API, 兔子API, 兔子API, 180txt API. Check each provider row before using it because free tier limits can change.