VNGRS

Kumru 7B

otherReleased Apr 2025

Turkish large language model developed by VNGRS, pre-trained from scratch on 500 GB of Turkish corpora (300B tokens). Decoder-only architecture with a custom tokenizer optimized for Turkish, supporting code, math, and chat. Outperforms significantly larger multilingual models on the Cetvel Turkish benchmark. Available for on-premise enterprise deployment.

Capabilities

chatcompletioncode-generation

Modalities

textcode

Context Window

8K tokens

Providers

available

Available from 2 providers

Cheapest

Featherless

$0.00/1M tokens

Featherless, Hugging Face Inference

Providers (2)

Sorted by total cost (input + output per 1M tokens). Click a row to view provider details.

Provider	Pricing (per 1M)	Rate Limits	Regions	Health	Latency
Featherless	In: FreeOut: Free	60 RPM / 300K TPM	global	Healthy	0ms
Hugging Face Inference	In: FreeOut: Free	60 RPM / 300K TPM	us-east-1eu-west-1	Healthy	0ms

Quick Start

Use this model via Featherless with an OpenAI-compatible SDK.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.featherless.com/v1",
  apiKey: process.env.FEATHERLESS_API_KEY,
});

const response = await client.chat.completions.create({
  model: "vngrs-ai/Kumru-7B",
  messages: [
    { role: "user", content: "Hello!" }
  ],
});

console.log(response.choices[0].message.content);

Using Featherless API • OpenAI-compatible SDK