GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
GoDotWebs
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
GoDotWebs
Model Tree
Input Modalities
Output Modalities
Supported Functionality
https://godotwebs.com/dotwebs-1
Our frontier-class model. Trained on mimicked GoDotWebs practical use cases and equipped with a built-in reliability architecture that actively minimizes hallucinations for more accurate, trustworthy results.
DotWebs 1 has been trained on an extensive dataset comprising thousands of mimicked real-world applications, mirroring the expertise gained by a human who has authored and optimized thousands of such documents. The purpose of the model is to make the application process significantly faster and smoother, while delivering feedback that is far more realistic, precise, and personalized, moving beyond generic responses.
We evaluated DotWebs 1 using two benchmarks: Application Evaluation (AE), our in-house assessment designed around GoDotWebs workflows based on successful Y-combinator applications.
In each run, the applicant model receives the company profile and drafts an answer. The judge model then scores that output against the original application that earned the company a spot in Y Combinator. Scoring breaks down into two criteria: Format Score and Closeness Score. Format Score measures how closely the answer follows the GoDotWebs workflow; Closeness Score measures how closely it matches the original model answer.
For the judge model, we chose a cost-balanced option.
Judge Model: openai/gpt-oss-20b
Formula: composite = formatScore * 0.7 + closenessScore * 0.3
| Models | Scores |
|---|---|
| GoDotWebs/DotWebs-1 | 66.00% |
| Google/Gemma-3n-E4B-it | 58.48% |
| Nvidia/Nemotron-3-Ultra-550B-a55b | 57.47% |
| Qwen/Qwen3.5-9B | 52.10% |
| Llama-3.3-70B-Instruct-Turbo | 51.38% |
Model Name: GoDotWebs/DotWebs-1
Base Model: Qwen/Qwen3.5-9B
Primary Usage: Form auto-filling, answers evaluation, feedback simulations.
DotWebs 1 gives you the option to run on our own model instead of relying on third-party providers. Your data stays within our platform, protected by layered security controls designed to keep applications and personal information safe.
Our roadmap focuses on scaling DotWebs 1 or upcoming models to meet growing demand while advancing faster inference, higher accuracy, and more realistic outputs tailored to GoDotWebs applications.
DotWebs 1 is made available through GoDotWebs and is intended for use within our platform. GoDotWebs retains rights to our branding, custom model work, and related materials. Your use is governed by your plan and our Terms of Service. For licensing questions, contact info@godotwebs.com.