Lebrel Brings Uncensored DeepSeek V4.1 Flash From $0.13 per Million Input Tokens, Adds Alibaba's Qwen4-Max on Day One

Built on DeepSeek V4.1 Flash 1017, the edition has no added refusal layer. Lebrel also offers nine routed models through one API key and balance, now including Alibaba’s newly released Qwen4-Max.


Lebrel Offers Uncensored DeepSeek V4.1 Flash From $0.13 per Million Input Tokens, Adds Qwen4 Max on Day One

DOVER, Del., Nov. 18, 2026 (GLOBE NEWSWIRE) --

Lebrel, an AI inference platform, offers its own uncensored edition of DeepSeek V4.1 Flash in Chat and through an OpenAI-compatible API. Built on the 1017 release, the edition gives users access without an added refusal layer, with pricing from $0.13 per million input tokens and $0.53 per million output tokens.

The platform combines the uncensored edition with nine models routed from DeepSeek, Moonshot, Alibaba and Z.ai, including Qwen4-Max, which Lebrel added on the day Alibaba released it. Users can access both offerings through one API key and balance, with a signed receipt accompanying every answer.

Uncensored access built on a leading agent model

Lebrel’s edition is built on DeepSeek V4.1 Flash 1017, a base model with strong published results across software engineering and agent tasks. Lebrel serves the edition without adding a refusal layer on top, allowing users to explore its capabilities directly in Chat or integrate it into applications and agents.

In DeepSeek’s published comparison, V4.1 Flash 1017 leads the listed Claude Opus 5, GPT-5.6 Sol, GLM-5.3 and Kimi K3 models on the following evaluations:

  • DeepSWE v1.1: 77.0
  • AutomationBench: 60.5
  • Agents’ Last Exam: 35.4
  • Apex Bench v1.1: 72.1

Lebrel Offers Uncensored DeepSeek V4.1 Flash From $0.13 per Million Input Tokens, Adds Qwen4 Max on Day One 2

These are DeepSeek’s published base-model scores under its evaluation settings. They were not measured by Lebrel and are not a separate evaluation of Lebrel’s uncensored edition.

Low-cost inference in Chat and through the API

The uncensored edition starts at $0.13 per million input tokens and $0.53 per million output tokens during lower demand. Standard rates are $0.33 input and $1.32 output per million tokens. Current rates and optional priority pricing are available at lebrel.ai/pricing.

Developers can connect through an OpenAI-compatible API, making it possible to use Lebrel within applications and agent workflows through a familiar interface.

Nine routed models, one key

The same API key and balance also provide access to nine routed models: Qwen4-Max, DeepSeek V4.1 Flash, DeepSeek V4 Pro, Kimi K3.1 Pro, Kimi K3, GLM-5.3, GLM-5.3 Flash, Qwen3.8 Max and Qwen3.8 Flash.

The newest addition is Qwen4-Max, which Lebrel added on the day Alibaba released it. In Qwen’s published comparison, Qwen4-Max ranks first among the listed models on DeepSWE v1.1 (75.7), Agents’ Last Exam v1.1 (46.7), HLAB-Vals-1.1 (21.3) and BioMysteryBench (85.0), ahead of Kimi K3.1, GPT6.1 Sol, GPT6 Astra and Opus5.5. These are Qwen’s published scores and were not measured by Lebrel. Lebrel offers Qwen4-Max at $1.40 per million input tokens and $4.20 per million output tokens, nearly 7% below Alibaba’s list price of $1.50 and $4.50.

These models run through Lebrel’s attested router in Intel TDX (Trust Domain Extensions). The model provider sees the request, not who sent it. Each routed model is charged at its listed rate.

Lebrel’s own uncensored edition uses a separate serving path, with input encrypted from the user’s device to the model.

A signed receipt with every answer

Every answer comes with a signed receipt that users can inspect to check the edition or model route associated with their response. Receipts can be verified in the browser at lebrel.ai/verify.

This gives users an inspectable record of how an answer was served. It does not certify the factual accuracy of the answer.

Availability

Lebrel’s uncensored DeepSeek edition is available in Chat and through the API, with free access every day from 15:00 to 19:00 UTC during the open beta, subject to availability and usage limits. Paid access is available outside the free window. Routed models remain available at their listed rates.

Users can try the platform at lebrel.ai. Integration documentation is available at lebrel.ai/docs, and published edition records can be explored at models.lebrel.ai.

About Lebrel

Lebrel is an AI inference platform offering its own uncensored editions of open-weight models alongside models routed from external providers. It serves individuals, developers and agents through Chat and an OpenAI-compatible API, with a focus on affordable inference and inspectable records behind responses.

Website: https://lebrel.ai

 

Contact Data

GlobeNewswire