llm-benchmarks
githubdrose.io

parasail Provider Benchmarks


Measured over the last 30 days

Models Tracked
32
Avg Tokens / Second
27.95
Avg Time to First Token (ms)
3681.87
Last Updated
Oct 17, 2026

All parasail models

32 live · one row per model · fastest first · 2 folded behind their current lane
ModelAvg Toks/SecMinMaxAvg TTF (ms)
Qwen3 Next 80B A3B Instruct65.1065.1065.10560.00
MythoMax 13B55.9050.5061.30600.00
Llama 3 8B Lunaris49.8049.8049.80600.00
Cydonia 24B V4.148.2048.2048.20410.00
GLM 5.247.9047.3048.40880.00
Skyfall 36B V247.8047.8047.80540.00
Mistral Small 3.2 24B46.2046.2046.20490.00
Llama 3.2 3B Instruct43.9043.9043.90720.00
Mistral Nemo42.7029.4056.00850.00
Llama 4 Maverick37.7031.8043.60580.00
Qwen3 VL 235B A22B Instruct37.2037.2037.20660.00
Llama 3.3 70B Instruct36.9022.0051.20900.00
MiniMax M331.2015.2046.001870.00
Gemma 3 27B30.4030.4030.40870.00
gpt-oss-120b29.9015.3047.002040.00
Qwen3 Coder Next28.2028.2028.201250.00
Qwen3 235B A22B Instruct 250724.4016.6036.10780.00
Qwen2.5 VL 72B Instruct22.3022.3022.301200.00
UnslopNemo 12B22.3022.3022.30630.00
UI-TARS 7B20.8020.8020.80710.00
Gemma 4 31B20.5016.6024.40720.00
DeepSeek V4 Flash 073117.5013.7021.203060.00
Gemma 4 26B A4B17.305.5130.601390.00
gpt-oss-20b12.904.3920.406850.00
DeepSeek V4.1 Flash11.7011.7011.705330.00
Kimi K2.69.882.2440.209680.00
Qwen3.5-35B-A3B8.858.858.856890.00
Qwen3.5-9B7.287.287.288460.00
DeepSeek V4 Flash 04237.003.2414.8010400.00
DeepSeek V4 Pro 04235.395.395.3910300.00
Qwen3.6 35B A3B3.722.296.2917900.00
Qwen3.5 397B A17B3.651.667.3719700.00

Frequently Asked Questions

Which parasail model is fastest?
Based on recent tests, Qwen3 Next 80B A3B Instruct shows the highest average throughput among tracked parasail models.
How many recent measurements feed this dashboard?
This provider summary aggregates 3308 individual prompts measured across 2876 monitoring runs over the past month.