Uncensored IBM Models

Match · Updated July 2026

The best uncensored IBM model is Granite 4.1 8B (6.7 intelligence) — served without a provider-level moderation layer.

6.7Intelligence

$0.050Input /M

131KContext

Large language models served without a provider-level moderation layer, ranked by intelligence. "Uncensored" here means the API endpoint doesn't add its own filtering on top of the model — each model's own alignment training still applies, and open-weight entries can be self-hosted for full control. Relevant for fiction, roleplay, red-teaming and safety research, where an extra moderation layer causes over-refusal on legitimate prompts.

1
granite-4.1-8b
ToolsJSON6.7 intel · $0.050/M · 131K ctx
6.7
Intelligence

Frequently asked

What is the best uncensored IBM model?

The best uncensored IBM model is Granite 4.1 8B (6.7 intelligence) — served without a provider-level moderation layer.

How many IBM models are there?

modelgrep tracks 2 IBM models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by Granite 4.1 8B. 1 of them qualify for this ranking.

More IBM rankings

IBM: Smartest LLMs IBM: Best LLMs for Coding IBM: Best LLMs for Design & Frontend IBM: Fastest LLMs IBM: Lowest-Latency LLMs IBM: Cheapest LLMs IBM: Best Free LLMs IBM: Best Reasoning LLMs IBM: Best Vision LLMs IBM: Best LLMs for Agents IBM: Best Open-Source LLMs IBM: Longest-Context LLMs IBM: Best LLMs for Writing IBM: Best LLMs for Math & Science IBM: Best LLMs for RAG IBM: Best LLMs for SQL & Data Analysis IBM: Best LLMs for Roleplay

All rankings

Small & Fast LLMs Best Local LLMs Smartest LLMs Best LLMs for Coding Best LLMs for Design & Frontend Fastest LLMs Lowest-Latency LLMs Cheapest LLMs Best Free LLMs Best Reasoning LLMs Best Vision LLMs Best LLMs for Agents Best Open-Source LLMs Longest-Context LLMs Best LLMs for Writing Best LLMs for Math & Science Best LLMs for RAG Best LLMs for SQL & Data Analysis Best LLMs for Roleplay