How much does the OpenAI GPT-5.4 API cost?

GPT-5.4 API pricing is $2.50 per million input tokens and $15.00 per million output tokens. Use our calculator at aiapicost.com for exact cost estimates based on your usage.

Which AI model is cheapest for API usage?

The cheapest AI API models change frequently. Use aiapicost.com to compare real-time pricing across 400+ models from OpenAI, Anthropic, Google, DeepSeek, and more. DeepSeek and open-source models typically offer the lowest per-token costs.

How do AI API token costs work?

AI APIs charge per token (roughly 0.75 words). Costs are split into input tokens (what you send) and output tokens (what the model generates). Output tokens are typically 2-5x more expensive. Prices are quoted per 1 million tokens.

Claude vs ChatGPT: which is better?

Both are top-tier models. Claude excels at coding and instruction-following, while GPT-5.4 offers broader multimodal capabilities. Compare them head-to-head at aiapicost.com/compare with real benchmark data.

Live speed data from Artificial Analysis API

AI Model Speed Rankings

Compare 511+ AI models by response speed, latency, and throughput. Find the fastest models for your use case.

511 models · click headers to sort

Model

Throughput ↓

TTFT

$/1M

$/Speed

Price×TTFT

Mercury 2

Inception

818 t/s

4.90s

$0.38

$0.000

1837.9

Granite 3.3 8B (Non-reasoning)

IBM

426 t/s

21.98s

$0.09

$0.000

1868.0

Granite 4.0 H Small

IBM

364 t/s

8.74s

$0.11

$0.000

935.5

Nova Micro

Amazon

319 t/s

663ms

$0.06

$0.000

40.4

Qwen3.5 2B (Non-reasoning)

Alibaba

318 t/s

223ms

$0.04

$0.000

8.9

Nemotron 3 Nano Omni 30B A3B Reasoning

NVIDIA

307 t/s

585ms

$0.13

$0.000

76.6

Llama 3.1 Nemotron Instruct 70B

NVIDIA

295 t/s

235ms

$1.2

$0.004

282.0

Gemini 3.1 Flash-Lite Preview

Google

288 t/s

4.82s

$0.56

$0.002

2713.7

gpt-oss-120b (low)

OpenAI

276 t/s

509ms

$0.26

$0.001

133.4

gpt-oss-120b (high)

OpenAI

272 t/s

524ms

$0.26

$0.001

137.3

gpt-oss-20B (high)

OpenAI

270 t/s

363ms

$0.09

$0.000

31.9

gpt-oss-20B (low)

OpenAI

264 t/s

462ms

$0.10

$0.000

43.9

Gemini 2.5 Flash-Lite (Reasoning)

Google

238 t/s

19.33s

$0.17

$0.001

3382.4

Qwen3.5 Omni Flash

Alibaba

232 t/s

1.16s

$0.28

$0.001

318.4

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)

NVIDIA

230 t/s

687ms

$0.30

$0.001

206.1

Gemini 3.5 Flash (high)

Google

227 t/s

9.75s

$3.4

$0.015

32892.8

Qwen3 0.6B (Reasoning)

Alibaba

225 t/s

1.01s

$0.40

$0.002

403.6

Nova 2.0 Lite (Non-reasoning)

Amazon

223 t/s

838ms

$0.85

$0.004

712.3

Qwen3 0.6B (Non-reasoning)

Alibaba

223 t/s

893ms

$0.19

$0.001

167.9

Gemini 2.5 Flash (Reasoning)

Google

218 t/s

12.92s

$0.85

$0.004

10981.1

Gemini 2.5 Flash-Lite (Non-reasoning)

Google

203 t/s

2.86s

$0.17

$0.001

500.5

Nova Lite

Amazon

202 t/s

634ms

$0.10

$0.001

66.6

Gemini 3 Flash Preview (Reasoning)

Google

196 t/s

6.17s

$1.1

$0.006

6946.9

Llama 3.1 Instruct 8B

Meta

194 t/s

469ms

$0.10

$0.001

46.9

Gemini 2.5 Flash (Non-reasoning)

Google

193 t/s

572ms

$0.85

$0.004

486.2

Step 3.5 Flash

StepFun

193 t/s

905ms

$0.15

$0.001

135.8

Devstral Small (Jul '25)

Mistral

192 t/s

414ms

$0.15

$0.001

62.1

Step 3.5 Flash 2603

StepFun

187 t/s

868ms

$0.00

—

Jamba 1.6 Mini

AI21 Labs

187 t/s

758ms

$0.25

$0.001

189.5

Gemini 3 Flash Preview (Non-reasoning)

Google

186 t/s

878ms

$1.1

$0.006

987.8

Ministral 3 3B

Mistral

180 t/s

339ms

$0.10

$0.001

33.9

GPT-5.1 Codex mini (high)

OpenAI

177 t/s

5.33s

$0.69

$0.004

3667.0

Sarvam 30B (high)

Sarvam

177 t/s

1.15s

$0.00

—

Qwen3.6 35B A3B (Reasoning)

Alibaba

176 t/s

1.46s

$0.56

$0.003

810.4

Nova 2.0 Lite (high)

Amazon

175 t/s

10.15s

$0.85

$0.005

8631.7

NVIDIA Nemotron 3 Super 120B A12B (Reasoning)

NVIDIA

173 t/s

959ms

$0.41

$0.002

395.1

Qwen3.5 4B (Non-reasoning)

Alibaba

173 t/s

233ms

$0.06

$0.000

14.0

Nova 2.0 Lite (medium)

Amazon

173 t/s

19.51s

$0.85

$0.005

16586.9

Nova 2.0 Pro Preview (low)

Amazon

172 t/s

3.14s

$3.4

$0.020

10798.8

GPT-5 (ChatGPT)

OpenAI

170 t/s

550ms

$3.4

$0.020

1890.9

GPT-5.1 Codex (high)

OpenAI

170 t/s

3.60s

$3.4

$0.020

12380.2

o4-mini (high)

OpenAI

169 t/s

17.80s

$1.9

$0.011

34265.0

GPT-5 Codex (high)

OpenAI

166 t/s

7.06s

$3.4

$0.021

24262.0

Qwen3.6 35B A3B (Non-reasoning)

Alibaba

166 t/s

1.43s

$0.84

$0.005

1203.5

Qwen3.5 4B (Reasoning)

Alibaba

165 t/s

249ms

$0.06

$0.000

14.9

GPT-5 nano (high)

OpenAI

164 t/s

72.48s

$0.14

$0.001

10002.2

o3-mini (high)

OpenAI

162 t/s

21.93s

$1.9

$0.012

42205.6

o3-mini

OpenAI

162 t/s

7.29s

$1.9

$0.012

14027.5

Qwen3.5 122B A10B (Reasoning)

Alibaba

160 t/s

1.10s

$1.1

$0.007

1206.7

Nova 2.0 Lite (low)

Amazon

160 t/s

5.06s

$0.85

$0.005

4304.4

Qwen3 Next 80B A3B Instruct

Alibaba

158 t/s

1.19s

$0.88

$0.006

1040.4

GPT-5.4 mini (xhigh)

OpenAI

157 t/s

4.34s

$1.7

$0.011

7327.6

Mistral Medium 3.5

Mistral

155 t/s

799ms

$3.0

$0.019

2397.0

GPT-5.4 nano (xhigh)

OpenAI

155 t/s

3.13s

$0.46

$0.003

1449.7

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)

NVIDIA

155 t/s

708ms

$0.10

$0.001

68.0

GPT-5.4 mini (Non-Reasoning)

OpenAI

154 t/s

522ms

$1.7

$0.011

881.1

GPT-5 nano (minimal)

OpenAI

154 t/s

718ms

$0.14

$0.001

99.1

Nova 2.0 Pro Preview (medium)

Amazon

154 t/s

15.08s

$3.4

$0.022

51845.0

GPT-5.4 nano (Non-Reasoning)

OpenAI

154 t/s

525ms

$0.46

$0.003

243.1

Qwen3.5 122B A10B (Non-reasoning)

Alibaba

152 t/s

1.05s

$1.1

$0.007

1152.8

GPT-5 nano (medium)

OpenAI

151 t/s

37.69s

$0.14

$0.001

5200.5

Qwen3 Next 80B A3B (Reasoning)

Alibaba

151 t/s

1.15s

$1.9

$0.012

2160.0

GPT-5.4 nano (medium)

OpenAI

150 t/s

3.84s

$0.46

$0.003

1776.1

MiMo-V2-Flash (Feb 2026)

Xiaomi

150 t/s

1.45s

$0.15

$0.001

217.5

GPT-5.4 mini (medium)

OpenAI

150 t/s

5.74s

$1.7

$0.011

9689.1

Mistral Small 4 (Reasoning)

Mistral

149 t/s

578ms

$0.26

$0.002

151.4

GPT-4o (Nov '24)

OpenAI

149 t/s

483ms

$4.4

$0.029

2113.1

GPT-4.1 nano

OpenAI

148 t/s

416ms

$0.17

$0.001

72.8

MiMo-V2-Flash (Reasoning)

Xiaomi

148 t/s

1.33s

$0.15

$0.001

199.3

Qwen3 VL 8B Instruct

Alibaba

147 t/s

945ms

$0.31

$0.002

292.9

MiMo-V2-Flash (Non-reasoning)

Xiaomi

146 t/s

1.48s

$0.15

$0.001

221.5

Qwen3 30B A3B 2507 (Reasoning)

Alibaba

144 t/s

1.14s

$0.67

$0.005

768.6

LFM2 24B A2B

Liquid AI

142 t/s

294ms

$0.05

$0.000

15.3

Mistral Small (Feb '24)

Mistral

142 t/s

519ms

$1.5

$0.011

778.5

Mistral Small (Sep '24)

Mistral

141 t/s

516ms

$0.30

$0.002

154.8

Gemini 3 Pro Preview (high)

Google

141 t/s

25.18s

$4.5

$0.032

113314.5

Mistral Small 4 (Non-reasoning)

Mistral

140 t/s

524ms

$0.26

$0.002

137.3

Qwen3 1.7B (Non-reasoning)

Alibaba

139 t/s

946ms

$0.19

$0.001

177.8

Mistral Small 3

Mistral

139 t/s

548ms

$0.10

$0.001

57.0

Qwen3 1.7B (Reasoning)

Alibaba

139 t/s

958ms

$0.40

$0.003

381.3

Qwen3.5 35B A3B (Non-reasoning)

Alibaba

138 t/s

1.09s

$0.69

$0.005

750.6

Sarvam M (Reasoning)

Sarvam

138 t/s

1.18s

$0.00

—

NVIDIA Nemotron Nano 9B V2 (Non-reasoning)

NVIDIA

138 t/s

702ms

$0.09

$0.001

60.4

Mistral Small 3.1

Mistral

136 t/s

509ms

$0.14

$0.001

70.2

Qwen3 VL 8B (Reasoning)

Alibaba

135 t/s

1.17s

$0.66

$0.005

768.9

Gemini 3.1 Pro Preview

Google

135 t/s

19.85s

$4.5

$0.033

89316.0

Nova 2.0 Pro Preview (Non-reasoning)

Amazon

134 t/s

825ms

$3.4

$0.026

2836.3

Qwen3.5 0.8B (Non-reasoning)

Alibaba

132 t/s

230ms

$0.02

$0.000

4.6

Sarvam 105B (high)

Sarvam

131 t/s

1.26s

$0.00

—

Qwen3.5 35B A3B (Reasoning)

Alibaba

131 t/s

1.29s

$0.69

$0.005

888.9

Trinity Large Thinking

Arcee AI

128 t/s

602ms

$0.40

$0.003

237.8

Granite 4.1 8B

IBM

127 t/s

474ms

$0.06

$0.000

29.9

Tiny Aya Global

Cohere

126 t/s

296ms

$0.00

—

Qwen3 VL 30B A3B (Reasoning)

Alibaba

126 t/s

1.03s

$0.34

$0.003

348.8

Gemini 2.5 Pro

Google

125 t/s

21.89s

$3.4

$0.027

75275.0

Qwen3 30B A3B 2507 Instruct

Alibaba

125 t/s

1.06s

$0.21

$0.002

225.1

Hy3-preview (Reasoning)

Tencent

123 t/s

2.43s

$0.20

$0.002

486.2

Qwen3 VL 30B A3B Instruct

Alibaba

121 t/s

1.01s

$0.30

$0.002

304.5

Llama 4 Maverick

Speed Metrics Guide

Throughput (tokens/s)

Output generation speed in tokens per second. Higher is better.

Good: >50 t/s · Excellent: >100 t/s

Time to First Token (TTFT)

Delay before the first token appears. Lower is better.

Good: <500ms · Excellent: <200ms

Price/Performance

Cost efficiency ratios. Lower values indicate better value.

$/Speed: price per t/s · Price×TTFT: latency penalty

Compare pricing for all models side by side

Open AI API Cost Calculator →

AI Model Speed Rankings

Speed Metrics Guide

Tools

Guides

Comparisons