The Open-Source AI Stack
RSS
All models

Models · Compare

Granite 3.0 8B Instruct vs Claude 3.5 Haiku

Rows highlighted in warm gray are where the models differ. Numbers carry their as-of date and primary source.

Specs

Field A: Granite 3.0 8B Instruct B: Claude 3.5 Haiku
Released 2024-10-212024-11-04
Developer IBMAnthropic
Openness OpenProprietary
License Apache-2.0Proprietary
OSI-approved yesno
Data released nono
Training code nono
Architecture denseunknown
Total params 8.1B
Active params
Experts
Context window 4K200K
Attention gqaunknown
Position enc. ropeunknown
Pretraining tokens 12.0T
Post-training sft, rlhfrlhf, constitutional
Training hardware H100
$/M input $0.80
$/M output $4.00
Output tok/sec 0

Benchmarks

Missing scores render as not reported; never inferred. Bold highlights the leader per benchmark.

General reasoning

MMLU 65.8 2024-10-21
MMLU-Pro 34.5 2024-10-21 63.4 2026-05-21
GPQA-Diamond 33.8 2024-10-21 40.8 2026-05-21

Code

HumanEval 64.6 2024-10-21
SWE-Bench Verified 40.6 2024-10-22
LiveCodeBench 31.4 2026-05-21

Math

MATH 72.1 2026-05-21
AIME 2024 3.3 2026-05-21

Held-out / arena

IFEval 52.3 2024-10-21

Context · A

IBM's first Apache-licensed enterprise instruct model in the Granite 3 line, trained on 12T tokens on IBM's Blue Vela H100 supercluster. Targeted at function calling, RAG, and 12-language multilingual workloads with permissive Apache-2.0 weights.

Context · B

Announced October 22 2024 alongside the upgraded Claude 3.5 Sonnet and the computer-use beta; generally available in early November. Anthropic positioned it as matching Claude 3 Opus on many intelligence benchmarks at Haiku-tier speeds, with a SWE-Bench Verified score of 40.6% making it briefly the best small-tier coding model from a major lab.

Granite 3.0 8B Instruct detail → · Claude 3.5 Haiku detail →