IB
IBM
Language model · United States ·Oct 2025

Granite-4.0-H-Tiny

Language Open (Restricted)
7B
Parameters
21
Benchmarks

About Granite-4.0-H-Tiny

Granite-4.0-H-Tiny is an AI model developed by IBM, in the language category, released in 2025, made available as an open-weights model with 7B params.

On this page you'll find Granite-4.0-H-Tiny's full specifications, including 21 benchmark results. Review provider pricing and benchmark scores below, or compare Granite-4.0-H-Tiny head-to-head with other language models.

Links

Run Granite-4.0-H-Tiny locally — what to buy

Share this build ↗

Sized to what Granite-4.0-H-Tiny actually needs (~8k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.

Software support

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Community

Would you run Granite-4.0-H-Tiny again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

Granite-4.0-H-Tiny benchmark scores
Benchmark Category Score Variant
HumanEval code 83% pass@1
HumanEval+ code 76% pass@1
MBPP code 80% pass@1
MBPP+ code 69% pass@1
BigCodeBench coding 41.06 pass@1
MultiPL-E coding 55.83 pass@1-multiple
AlpacaEval 2 general 30.61%
ArenaHard general 35.75%
IFEval general 84.78% instruct-strict
IFEval general 81.44% average
IFEval general 78.1% prompt-strict
MMLU general 68.65% 5-shot
MMLU-Pro general 44.94% 5-shot-cot
MMMLU general 61.87% 5-shot
AGIEval knowledge 62.15 0-shot-cot
INCLUDE knowledge 53.12 5-shot
GSM8K math 84.69% 8-shot
MGSM math 45.36 8-shot
BIG-Bench Hard reasoning 66.34% 3-shot-cot
GPQA Diamond reasoning 32.59% 0-shot-cot
BFCL v3 tool-use 57.65%

More models like Granite-4.0-H-Tiny

Frequently asked questions

What is Granite-4.0-H-Tiny?

Granite-4.0-H-Tiny is an AI model developed by IBM, in the language category, released in 2025. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created Granite-4.0-H-Tiny?

Granite-4.0-H-Tiny was developed by IBM and released in 2025.

Is Granite-4.0-H-Tiny open source or proprietary?

Granite-4.0-H-Tiny ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.

How much does Granite-4.0-H-Tiny cost?

Pricing for Granite-4.0-H-Tiny depends on the provider. See the providers table on this page for the latest API rates.

How does Granite-4.0-H-Tiny perform on benchmarks?

Granite-4.0-H-Tiny is benchmarked across 21 evaluations on GenAIList, including HumanEval (83%). See the full benchmark table below and compare it with other models.

No videos yet. Know a good one? Add it below.

Know a great video about Granite-4.0-H-Tiny?