Granite 4.0 Micro
IBMOpen WeightApache 2.0 · Commercial OK
Description
A preliminary version of the smallest model in the upcoming Granite 4.0 family, released May 2025. It utilizes a novel hybrid Mamba-2/Transformer, fine-grained mixture of experts (MoE) architecture (7B total parameters, 1B active at inference). This preview version is partially trained (2.5T tokens) but demonstrates significant memory efficiency and performance potential, validated for at least 128K context length without positional encoding.
Release Date
2025-09-22
Parameters
7.0B
Context Length
131K
Modalities
text
Capability Radar
15
general
17
coding
11
reasoning
20
scienceest.
13
agents
0
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 454 | 9.0 | AA |
| General Ranking | 473 | 16.0 | AA |
| Math Reasoning | 343 | 6.0 | AA |
| Reasoning | 37 | 78.0 | LS |
| Science | 448 | 18.0 | AA |
Benchmark Scores (LLM Stats)
Code
HumanEval
82.4%SR
Creativity
AlpacaEval 2.0
35.2%SR
Arena Hard
26.7%SR
Finance
MMLU
60.4%SR
TruthfulQA
58.1%SR
General
IFEval
63.0%SR
PopQA
22.9%SR
Language
BIG-Bench Hard
55.7%SR
Math
GSM8k
70.1%SR
DROP
46.2%SR
Reasoning
HumanEval+
78.3%SR
Safety
AttaQ
86.1%SR
AA Evaluation Indices
Math Index6.0
Intelligence Index2.4
Mmlu Pro0.4
Gpqa0.3
Ifbench0.2
Livecodebench0.2
Tau20.1
Scicode0.1
Aime 250.1
Hle0.1
Lcr0.0
Terminalbench Hard0.0
LLM Stats Category Scores
Safety90
Code80
Instruction Following60
Language60
Legal60
Math60
Structured Output60
Finance60
Healthcare60
Reasoning50
General40
Creativity30
Writing30
Pricing
Input PriceFree
Output PriceFree
Blended Price (3:1)Free
Speed
Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s
Provider Price Ranking
Provider Price Ranking
4 providers
Cheapest: OpenRouterMost Expensive: Cloudflare Workers AI
ProviderInputOutput
1OpenRouterCheapest
$0.017
$0.112
2Kilo Gateway
$0.017
$0.11
3Cloudflare AI Gateway
$0.017
$0.11
4Cloudflare Workers AI
$0.017
$0.112
Compare pricing across different API providers for this model.