SWT-API

SWT-API provides access to various AI models including Claude, Kimi, GLM, and Gemini for chat and completion tasks.

SWT-API offers 513 LLM API models.

API pricing per token ranges from $0.0000 to $0.375/M (input).

Speed benchmark average: 110 tok/s.

SWT-API is an API aggregator, offering models from multiple vendors.

Avg Speed110.15 tok/s

Latency4.66 s

Total Tests165

Models24

Updated4/17/2026

Created At12/7/2025

Supported Models

Model	Speed	Latency	Tests
qwen-3-32b-turbo	515.39 tok/s	2.33s	5
qwen-3-235b-a22b-instruct-2507	271.04 tok/s	1.09s	10
gemini-2.5-flash	139.00 tok/s	10.21s	15
sonoma-sky-alpha	133.73 tok/s	4.53s	10
gemini-2.5-flash-nothinking	132.03 tok/s	5.66s	15
deepseek-v3-0324-turbo	104.67 tok/s	1.19s	10
gemini-3-pro-poe	103.53 tok/s	13.75s	5
gemini-2.5-pro	102.41 tok/s	16.42s	5
chatgpt-4o-latest	89.21 tok/s	1.13s	5
gpt-5-mini-poe	88.90 tok/s	6.96s	5
gpt-4.1-nano	76.05 tok/s	2.16s	5
gpt-5-chat-poixe	73.29 tok/s	1.75s	5
gpt-5-chat-poe	69.39 tok/s	1.72s	15
gpt-5-poe	65.96 tok/s	2.24s	5
gpt-5-chat-latest	61.45 tok/s	0.99s	5
gpt-4o-search-preview-2025-03-11	60.84 tok/s	3.47s	5
gpt-4.1-mini	58.87 tok/s	1.82s	5
gpt-4.1	57.95 tok/s	1.20s	5
gpt-4.1-mini-2025-04-14	54.28 tok/s	1.11s	5
qwen3-235b-a22b	54.26 tok/s	3.20s	5

Showing 20 of 24 models.