Granite 4.0 350M

Granite 4.0 350M

IBM · Released Oct 28, 2025 Emerging
Intelligence #503 / 525
7.1 Our Score
AA Index #349 / 354
6.1 Artificial Analysis
Input
Not priced
Output
Not priced
Context
Not reported

Analysis Summary

Granite 4.0 350M sits in the Emerging tier on our leaderboard, ranked #503 of 525 published models on overall intelligence. At $0.000 input and $0.000 output per 1M tokens, it is among the most expensive on the market.

Editorial notes

Granite 4.0 350M from IBM is an ultra-compact model with near-zero coding and very low general reasoning scores, viable only for the most constrained embedded use cases.

Assessed April 26, 2026

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence1.3Technical0.6Value0Content1
Intelligence 1.3/10
Technical 0.6/10
Content 1/10
Value 0/10

Performance Indices

Source: Artificial Analysis

6.1 Intelligence Index
0.3 Coding Index
13.2 Agentic Index

Benchmark Scores

Intelligence

GPQA Diamond 26.1% Graduate-level scientific reasoning
HLE 5.7% Humanity's Last Exam
MMLU Pro 12.4% Multi-task language understanding
SciCode 0.9% Scientific computing

Technical

LiveCodeBench 2.4% Live coding evaluation
τ²-Bench 13.2% Conversational agent benchmark

Content

IFBench 15.9% Instruction following

Benchmark data from Artificial Analysis and Hugging Face

How does Granite 4.0 350M stack up?

Compare side-by-side with other emerging models.

Compare Models

Model Information

ProviderIBM
Release Date October 28, 2025
Status Active