Pa Providers· Inference host· 34 tracked models

Parasail — pricing & performance

Compare models All providers Data sources

Models served

of 908 tracked models

Mean output speed

tokens per second, averaged over
32 measured endpoints

Mean latency

0.9s

time to first token, averaged over
32 measured endpoints

Cheapest blended

$0.060

per 1M tokens at a 3:1
input:output blend

Output speed on Parasail

Median output tokens per second for each model this host serves, derived from OpenRouter endpoint telemetry. Higher is better. We do not run these measurements ourselves.

12 of 34 models ▾ ☰⚙⇩ +Add model from specific provider

207

141

108

gpt-oss-120b

Qwen3.5-35B-A3B

Llama 3.2 3B
Instruct

MiniMax M2.5

Qwen3 Next 80B A3B
Instruct

Qwen3.5 397B A17B

GLM 5.1

Kimi K2.7 Code

Trinity Large
Thinking

Llama 4 Maverick

Skyfall 36B V2

MiniMax M3

The Known Good

34 models on Parasail

Full leaderboard →

Model	Creator	Input $/M	Output $/M	Blended	Speed	TTFT	Context
AA Arcee AI: Trinity Large Thinking	arcee-ai	$0.22	$0.85	$0.38	67	0.5s	262k
ByteDance: UI-TARS 7B	ByteDance	$0.10	$0.20	$0.12	11	1.4s	128k
DeepSeek: DeepSeek V4 Flash	DeepSeek	$0.14	$0.28	$0.17	52	0.8s	1.05M
DeepSeek: DeepSeek V4 Pro	DeepSeek	$1.74	$3.48	$2.17	—	—	1.05M
Google: Gemma 3 27B	Google	$0.080	$0.45	$0.17	54	0.5s	131k
Google: Gemma 4 26B A4B	Google	$0.13	$0.40	$0.20	39	0.6s	262k
Google: Gemma 4 31B	Google	$0.15	$0.40	$0.21	21	1.6s	262k
Meta: Llama 3.2 3B Instruct	Meta	$0.050	$0.33	$0.12	108	0.3s	131k
Meta: Llama 3.3 70B Instruct	Meta	$0.22	$0.50	$0.29	24	1.5s	131k
Meta: Llama 4 Maverick	Meta	$0.35	$1.00	$0.51	58	0.4s	524k
MiniMax: MiniMax M2.5	MiniMax	$0.30	$1.20	$0.53	95	0.6s	197k
MiniMax: MiniMax M3	MiniMax	$0.30	$1.20	$0.53	56	0.7s	1.05M
Mistral: Mistral Small 3.2 24B	Mistral	$0.090	$0.30	$0.14	—	—	131k
MoonshotAI: Kimi K2.6	Moonshot AI	$0.75	$3.50	$1.44	42	0.8s	262k
MoonshotAI: Kimi K2.7 Code	Moonshot AI	$0.76	$3.50	$1.45	70	0.6s	262k
OpenAI: gpt-oss-120b	OpenAI	$0.10	$0.75	$0.26	207	0.3s	131k
OpenAI: gpt-oss-20b	OpenAI	$0.030	$0.15	$0.060	50	0.8s	131k
Qwen: Qwen2.5 VL 72B Instruct	Alibaba	$0.80	$1.00	$0.85	22	1.6s	128k
Qwen: Qwen3 235B A22B Instruct 2507	Alibaba	$0.14	$0.80	$0.30	24	0.6s	131k
Qwen: Qwen3 Coder Next	Alibaba	$0.12	$0.80	$0.29	56	0.9s	262k
Qwen: Qwen3 Next 80B A3B Instruct	Alibaba	$0.10	$1.10	$0.35	95	0.5s	262k
Qwen: Qwen3 VL 235B A22B Instruct	Alibaba	$0.21	$1.90	$0.63	3	0.9s	131k
Qwen: Qwen3 VL 8B Instruct	Alibaba	$0.25	$0.75	$0.38	26	0.9s	262k
Qwen: Qwen3.5 397B A17B	Alibaba	$0.50	$3.60	$1.27	80	0.4s	262k
Qwen: Qwen3.5-35B-A3B	Alibaba	$0.15	$1.00	$0.36	141	0.4s	262k
Qwen: Qwen3.5-9B	Alibaba	$0.10	$0.25	$0.14	27	0.6s	262k
Qwen: Qwen3.6 35B A3B	Alibaba	$0.15	$1.00	$0.36	55	1.1s	262k
Th TheDrummer: Cydonia 24B V4.1	thedrummer	$0.30	$0.50	$0.35	39	0.4s	131k
Th TheDrummer: Rocinante 12B	thedrummer	$0.25	$0.50	$0.31	50	0.4s	66k
Th TheDrummer: Skyfall 36B V2	thedrummer	$0.55	$0.80	$0.61	57	0.2s	33k
Mi Xiaomi: MiMo-V2.5	Xiaomi	$0.14	$0.28	$0.17	30	1.0s	1.05M
Z.ai: GLM 5	Z.ai	$1.00	$3.20	$1.55	42	0.8s	203k
Z.ai: GLM 5.1	Z.ai	$1.40	$4.40	$2.15	71	1.0s	203k
Z.ai: GLM 5.2	Z.ai	$1.40	$4.40	$2.15	45	5.3s	262k

The Known Good

Source. Prices, endpoint availability and throughput for Parasail are ingested from OpenRouter; model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint — it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this site; see methodology. Quality scores are not provider-specific: a model scores the same wherever it is hosted, so quality lives on the model pages.