Crof model pricing.

prices as of · 18 models · 31 prices

Serves open-weight models at Q8_0, int4 and mxfp4 precision and says so per model.

Crof is a host selling 18 models, from $0.27 per million tokens for Qwen3.5-9B, read on 16 Sept 2026.

Crof website

Kind of provider

Host

home country not stated

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 Crof's prices land, and why

Crof publishes 31 prices we can compare, and 31 of them are below the list price for that model. 29 of those have a published reason behind them, so they rank like any other price. Nobody here has read its entity, terms yet, so treat what it says about itself as its own claim.

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

not yet checked: entity, terms

Models to start with

all 31 prices ↓

Everything Crof 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

29 prices

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

  • Qwen3.5-9B

    Alibaba Cloud · 262k context

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

    $0.04 / $0.15in / out per M

    0.60×of list price

  • Qwen3.5-9B

    Alibaba Cloud · 262k context

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

    $0.04 / $0.15in / out per M

    0.60×of list price

  • DeepSeek V4 Flash 0731

    DeepSeek · 1M context

    Serves the open weights at Q8_0, a lower precision than the maker's, so not quite the same product.

    $0.07 / $0.10in / out per M

    0.30×of list price

  • DeepSeek V4 Flash 0731

    DeepSeek · 1M context

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

    $0.08 / $0.10in / out per M

    0.32×of list price

  • GLM-5.3-Flash

    Z.ai · 1M context

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

    $0.07 / $0.22in / out per M

    0.91×of list price

  • GLM-5.3-Flash

    Z.ai · 1M context

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

    $0.07 / $0.22in / out per M

    0.91×of list price

  • DeepSeek V4 Flash Vision Exp

    DeepSeek · 1M context

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

    $0.08 / $0.20in / out per M

    0.42×of list price

  • DeepSeek V4 Flash Vision Exp

    DeepSeek · 1M context

    Serves the open weights at Q8_0, a lower precision than the maker's, so not quite the same product.

    $0.08 / $0.20in / out per M

    0.42×of list price

  • Qwen3.8 27B

    Alibaba Cloud · 1M context

    Serves the open weights at Q4_0, a lower precision than the maker's, so not quite the same product.

    $0.09 / $0.30in / out per M

    0.15×of list price

  • DeepSeek V4.1 Flash

    DeepSeek · 1M context

    Serves the open weights at Q8_0, a lower precision than the maker's, so not quite the same product.

    $0.10 / $0.50in / out per M

    0.76×of list price

  • DeepSeek V3.2

    DeepSeek · 164k context

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

    $0.18 / $0.35in / out per M

    0.72×of list price

  • DeepSeek V4 Pro

    DeepSeek · 1M context

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

    $0.35 / $0.80in / out per M

    0.85×of list price

  • DeepSeek V4 Pro

    DeepSeek · 1M context

    Serves the open weights at Q8_0, a lower precision than the maker's, so not quite the same product.

    $0.35 / $0.80in / out per M

    0.85×of list price

  • GLM-5.2

    Z.ai · 1M context

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

    $0.30 / $1.05in / out per M

    0.23×of list price

  • GLM-5.2

    Z.ai · 1M context

    Serves the open weights at Q8_0, a lower precision than the maker's, so not quite the same product.

    $0.30 / $1.05in / out per M

    0.23×of list price

  • MiMo-V2.5-Pro

    Xiaomi · 1M context

    8% under the list price, an ordinary reseller margin.

    $0.40 / $0.80in / out per M

    0.92×of list price

  • MiMo-V2.5-Pro

    Xiaomi · 1M context

    8% under the list price, an ordinary reseller margin.

    $0.40 / $0.80in / out per M

    0.92×of list price

  • Qwen3.6 27B

    Alibaba Cloud · 262k context

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

    $0.20 / $1.50in / out per M

    0.39×of list price

  • Qwen3.8 27B

    Alibaba Cloud · 1M context

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

    $0.20 / $1.50in / out per M

    0.55×of list price

  • GLM-5.3

    Z.ai · 1M context

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

    $0.40 / $1.40in / out per M

    0.30×of list price

  • GLM-5.3

    Z.ai · 1M context

    Serves the open weights at Q8_0, a lower precision than the maker's, so not quite the same product.

    $0.40 / $1.40in / out per M

    0.30×of list price

  • Qwen3.5 397B-A17B

    Alibaba Cloud · 262k context

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

    $0.35 / $1.75in / out per M

    0.52×of list price

  • Kimi K2.6

    Moonshot AI · 262k context

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

    $0.50 / $1.99in / out per M

    0.51×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.

    $0.50 / $1.99in / out per M

    0.51×of list price

  • GLM-5.1

    Z.ai · 200k context

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

    $0.45 / $2.15in / out per M

    0.41×of list price

  • Kimi K2.7 Code

    Moonshot AI · 262k context

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

    $0.55 / $2.25in / out per M

    0.57×of list price

  • Kimi K2.7 Code

    Moonshot AI · 262k context

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

    $0.55 / $2.25in / out per M

    0.57×of list price

  • Kimi K3

    Moonshot AI · 1M context

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

    $2 / $8in / out per M

    0.58×of list price

  • Kimi K3

    Moonshot AI · 1M context

    Serves the open weights at mxfp4, a lower precision than the maker's, so not quite the same product.

    $2 / $8in / out per M

    0.58×of list price

Same price as the maker

2 prices

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

  • Gemma 4 31B IT

    Google · 262k context

    11% above OpenRouter's price.

    $0.10 / $0.30in / out per M

    0.98×of list price

  • Gemma 4 31B IT

    Google · 262k context

    11% above OpenRouter's price.

    $0.10 / $0.30in / out per M

    0.98×of list price

Frequently asked

What is Crof?
Serves open-weight models at Q8_0, int4 and mxfp4 precision and says so per model. It sells 18 of the models we track.
Is Crof cheaper than buying from the maker?
31 of its 31 prices sit below the list price for the model, and 29 of those have a published reason behind them, so they can lead a ranking.
Does Crof train on what you send it?
Crof does not say whether it trains on what you send it. It does not say how long it keeps requests.
How current are these Crof prices?
They come from its own public catalog, last read 16 Sept 2026. Prices exclude tax.