Parasail model pricing.

prices as of · 30 models · 30 prices

Serves open-weight models on its own GPUs; batch at half the serverless price and FP16 at a 30% premium.

Parasail is a host selling 30 models, from $0.12 per million tokens for Mistral Nemo, read on 16 Sept 2026.

Parasail prices

Kind of provider

Host

based in US

Where it runs

not stated

no regions published

How you pay

not stated

no payment terms published

Free tier

none published

nothing given away

Where Parasail's prices land, and why

Parasail publishes 30 prices we can compare, and 13 of them are below the list price for that model. 14 of those have a published reason behind them, so they rank like any other price. Nobody here has read its terms yet, so treat what it says about itself as its own claim.

Parasail does not say whether it trains on what you send it. It does not say how long it keeps requests.

not yet checked: terms · privacy policy ↗

Models to start with

all 30 prices ↓

Everything Parasail sells, against the list price

Cheapest first inside each group. The share is this row against the price the lab that trained the model charges.

Cheaper, and we know why

14 prices

Below list, with a published reason: hosting the open weights, a regional price list, credits, a stated discount.

  • Mistral Nemo

    Mistral AI · 128k context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.03 / $0.03in / out per M

    0.20×of list price

  • Mistral Small 3.2 24B

    Mistral AI · 256k context

    4% under the list price, an ordinary reseller margin.

    $0.09 / $0.30in / out per M

    1.07×of list price

  • DeepSeek V4 Flash 0731

    DeepSeek · 1M context

    7% under the list price, an ordinary reseller margin.

    $0.14 / $0.28in / out per M

    0.67×of list price

  • Gemma 4 26B A4B IT

    Google · 262k context

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision bf16.

    $0.13 / $0.40in / out per M

    0.75×of list price

  • Qwen3 Coder Next

    Alibaba Cloud · 262k context

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision bf16.

    $0.12 / $0.80in / out per M

    0.48×of list price

  • Qwen3 235B-A22B

    Alibaba Cloud · 131k context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.14 / $0.80in / out per M

    0.25×of list price

  • Qwen3-Next 80B-A3B Instruct

    Alibaba Cloud · 131k context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.10 / $1.10in / out per M

    0.40×of list price

  • Qwen3.5 35B-A3B

    Alibaba Cloud · 262k context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.15 / $1in / out per M

    0.53×of list price

  • Qwen3.6 35B-A3B

    Alibaba Cloud · 262k context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.15 / $1in / out per M

    0.65×of list price

  • Qwen3-VL 235B-A22B

    Alibaba Cloud · 131k context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.21 / $1.90in / out per M

    0.52×of list price

  • Qwen3.8 27B

    Alibaba Cloud · 1M context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.24 / $2.20in / out per M

    0.76×of list price

  • Qwen2.5-VL 72B Instruct

    Alibaba Cloud · 131k context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.80 / $1in / out per M

    0.20×of list price

  • Qwen3.5 397B-A17B

    Alibaba Cloud · 262k context

    Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.

    $0.50 / $3.60in / out per M

    0.94×of list price

  • Kimi K2.6

    Moonshot AI · 262k context

    Serves the open weights at int4, a lower precision than the maker's, so not quite the same product.

    $0.75 / $3.50in / out per M

    0.84×of list price

Same price as the maker

16 prices

The same price the maker charges, sold by someone else.

Frequently asked

What is Parasail?
Serves open-weight models on its own GPUs; batch at half the serverless price and FP16 at a 30% premium. It sells 30 of the models we track, and trades as Parasail, based in US.
Is Parasail cheaper than buying from the maker?
13 of its 30 prices sit below the list price for the model, and 14 of those have a published reason behind them, so they can lead a ranking.
Does Parasail train on what you send it?
Parasail does not say whether it trains on what you send it. It does not say how long it keeps requests.
How current are these Parasail prices?
They come from the public catalogs we read, last read 16 Sept 2026. Prices exclude tax.