Skip to content
ModelCap

Inference provider

Parasail: AI models and API pricing

Every tracked model Parasail serves through OpenRouter, with Parasail's own input and output price per 1M tokens, the context window it serves, disclosed quantization and measured endpoint uptime. Rows link to the model page, where every provider's deployment is listed side by side.

Snapshot as of 11 August 2026

Parasail serves 25 tracked models through 25 endpoints as of 11 August 2026; the highest-ranked is GLM 5.2 (ModelCap Index #13); mean measured uptime is 98.86%.

Models served
25
On the ModelCap Index
20
Cheapest output price
$0.03/1M

Language models on Parasail (25)

Ranked models first, in ModelCap Index order. Prices are the lowest Parasail lists for the model; the endpoint count shows how many deployments carry it.

Language models served by Parasail, with price, context, quantization and uptime
#ModelModelCap IndexInput $/1MOutput $/1MContextQuantizationUptime (30 min)Endpoints
1GLM 5.2Z.ai#1370.5$1.40$4.40262Kfp499.26%1Open weights
2Gemma 4 31BGoogle#2062.8$0.15$0.40262Kfp899.62%1Open weights
3Qwen3.5 397B A17BQwen#2261.6$0.50$3.60262Kfp899.77%1Open weights
4MiniMax M3MiniMax#2361.3$0.30$1.201Mfp895.19%1
5Gemma 4 26B A4BGoogle#2559.2$0.13$0.40262Kbf1698.74%1Open weights
6Qwen3 235B A22B Instruct 2507Qwen#3153.6$0.14$0.80131Kfp899.89%1Open weights
7Qwen3 VL 235B A22B InstructQwen#3451.5$0.21$1.90131Kfp899.38%1Open weights
8Qwen3 Next 80B A3B InstructQwen#4046.6$0.10$1.10262Kfp8100.0%1Open weights
9Qwen3.6 35B A3BQwen#4146.3$0.15$1.00262Kfp899.94%1Open weights
10Trinity Large ThinkingArcee AI#6036.5$0.22$0.85262Kfp4100.0%1
11Gemma 3 27BGoogle#6533.9$0.08$0.45131Kfp899.76%1
12Mistral Small 3.2 24BMistral AI#6832.3$0.09$0.30131Kbf1699.74%1Open weights
13UI-TARS 7BByteDance#11025.5$0.10$0.20128Kbf161Open weights
14Llama 4 MaverickMeta#18123.5$0.35$1.00524Kfp8100.0%1
15Kimi K2.7 CodeMoonshot AI#18921.5$0.76$3.50262Kint498.99%1
16Qwen3 Coder NextQwen#19021.2$0.12$0.80262Kbf16100.0%1Open weights
17Llama 3.3 70B InstructMeta#19320.3$0.22$0.50131Kfp898.03%1
18Mistral NemoMistral AI#20311.9$0.03$0.03131Kfp8100.0%1Open weights
19Qwen3 VL 8B InstructQwen#2108.8$0.25$0.75262Kbf1695.27%1Open weights
20Llama 3.2 3B InstructMeta#2260.0$0.05$0.33131Kbf16100.0%1
21Qwen2.5 VL 72B InstructQwen$0.80$1.00128Kfp898.66%1
22Qwen3.5-9BQwen$0.10$0.25262Kbf16100.0%1Open weights
23GLM 5.1Z.ai$1.40$4.40203Kfp891.62%1Open weights
24Kimi K2.6Moonshot AI$0.75$3.50262Kint4100.0%1
25Qwen3.5-35B-A3BQwen$0.15$1.00262Kfp81Open weights

Parasail on ModelCap: common questions

How many AI models does Parasail serve?

25 tracked models as of 11 August 2026: 25 language models (20 with a public ModelCap Index position), across 25 live endpoints observed through OpenRouter.

What is the best model available on Parasail?

GLM 5.2 from Z.ai is the highest-ranked language model Parasail serves, at ModelCap Index position #13 and $4.40 per 1M output tokens on Parasail. The table lists every model in Index order.

What is the cheapest model on Parasail?

Mistral Nemo at $0.03 per 1M output tokens ($0.03 per 1M input), the lowest positive language-token price Parasail lists. Free-tier endpoints, where present, show as $0 and are excluded from that comparison.

Which Parasail model has the largest context window?

MiniMax M3, served with a 1M-token context window on Parasail. Providers sometimes serve less than the model's published maximum, so the figure here is the provider's own.

How reliable is Parasail?

Mean measured uptime across Parasail endpoints is 98.86% over the last 30 minutes, as reported by OpenRouter at the time of the snapshot. Per-model uptime is in the table; it is a live operational signal, not a service-level guarantee.

Where do the Parasail prices come from?

From the OpenRouter catalogue entry for each Parasail endpoint, in USD per 1M tokens, re-observed on every ModelCap refresh. Where Parasail exposes several deployments of one model at different prices, the table shows the lowest and notes the count; the model page lists each deployment.

How current is this page?

It re-renders within a minute of every dataset refresh; this snapshot was published 11 August 2026. Offers retained from a failed refresh are marked stale in the data and are not counted here.

Explore further