Granite 4 350M is IBM's language model. IBM's ultra-compact 350M-parameter Granite 4 model designed for efficient on-device and edge deployment in enterprise NLP tasks.
Specifications
Canonical IDibm-granite-4-350m
TypeLanguage
StatusActive
CreatorIBMIBM
Input ModalitiesText
Output ModalitiesText
Parameters0.35B
Benchmarks
Intelligence Index
#560
Math Index
#272
MMLU-Pro
#340
GPQA
#517
HLE
#355
LiveCodeBench
#330
IFBench
#424
Time to First Token
#167
SciCode
#543
AIME 2025
#272
LCR
#465
TerminalBench Hard
#388
TAU2
#372
Output TPS
#405

Capabilities

Input1/5
Text✓
Image·
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·

Cheapest Instances to Run It

Cloud GPU instances that can host Granite 4 350M, ranked by cheapest on-demand price. The model needs about 1 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.

All clouds
FP16 (full precision)
US Dollar ($)
Instance
Cloud
GPU
VRAM
Price
Cheapest region
g6f.largeAWSL43 GB$0.202/hr
Standard_NV4as_v4AzureAMD Radeon Instinct MI2516 GB$0.233/hr
g6f.xlargeAWSL43 GB$0.237/hr
7 more instances can run Granite 4 350M
Unlock the full ranked list and FP8 / INT4 quantization with a CloudPrice subscription.

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
Granite 4.2 8B131K$0.060$0.150Available
Granite 4.1 8B131K$0.050$0.100Deprecated
Granite 4 Micro H131K$0.017$0.112Available
Granite 4 350M————Current
Granite 4.2 30B—131K——Available
Granite 4.2 3B—131K——Available
Granite 4.1 30B————Available
Granite 4 Small H—131K$0.064$0.265Available
Granite 4.1 3B————Available
Granite 4 Nano 1B H————Available
Granite 4 Micro————Available

Model IDs

granite-4-0-350m
granite-4-0-h-350m
ibm-granite-4-350m