ModelPriceWatch.com
Last scan 2026-09-23 Models tracked 270 Providers 35 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

Granite 4 H Small

by IBM

Current budget open weights cheap tier
Email me when Granite 4 H Small gets cheaper
Availability: IBM's watsonx.ai rate card is the only per-token price this model has, and it is the only Granite 4 SKU IBM sells per token — granite-4h-micro and granite-4h-tiny are listed "Not available" on the same table. IBM prices the whole table at a uniform 6% over round base rates (granite-8b-code-instruct USD 0.636, granite-guardian-3-8b USD 0.212, llama-3-3-70b USD 0.7526), so $0.0636/$0.265 is the rate card figure, not a rounding of $0.06/$0.25. IBM notes its prices are indicative, may vary by country, and exclude taxes and duties.

Today's price · per 1M tokens

Input

$0.064

per 1M tokens

Output

$0.265

per 1M tokens

Blended

$0.114

blended $/1M — 3:1 weighted input:output

Source: official IBM pricing · read MODELPRICEWATCH.COM · 2026-09-23

Price receipt

We read IBM's own pricing page on and found granite-4h-small listed with a price on the same row — that read is where the number above comes from. We keep the page text we read; its content hash is 2a14968924.

Overview

IBM Granite 4 H Small model. $0.0636/$0.265 per 1M tokens.

Run it yourself

Deploy this open model on rented GPUs

Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.

RunPod →
Rent GPUs by the second and serve any open-weight model yourself.
Novita AI →
Serverless open-model inference billed per token — no GPU ops.
Vast.ai →
GPU marketplace — rent spot/on-demand GPUs cheaply and serve any open-weight model.

Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.

Capabilities

struck through = not supported
Input 1/4
Text ✓ Image Audio Video
Output
Text ✓
Features 2/9
Prompt caching Reasoning Coding Fast inference Long context ✓ Open weights ✓ Multimodal Web search Realtime

Benchmark performance

accuracy % · higher is better
Percentile vs all tracked models 3th1 independent measurement
Percentile per $/Mtok 27.3
MMMLU vendor
69.69%
BFCL vendor
64.69%
HumanEval vendor
88%

Every per-benchmark score we hold for this model is reported by its own vendor, so none is counted toward an accuracy average or any ranking. The percentile above is its standing across the independent composites below.

Independent composite scores
  • AA Intelligence Index: 5 (Artificial Analysis composite across reasoning, knowledge and coding evals)
How it stacks up
  • Ranks #181 of 187 comparably-measured models by percentile score, across 1 independent measurement
  • Ranks #104 of 187 comparably-tested models by normalized performance per dollar

414.5 tokens/sec output

Source: Artificial Analysis, IBM model card, IBM model card (5-shot) · updated · See full rankings →

Specifications

Provider
IBM
Context window
128K tokens
Modality
text
Parameters
Proprietary
Open source
Yes — open weights available
Released
Status
Current
Last updated
Tags
open-weightsbudget

Availability verified: — listed on IBM's own page

Try on IBM Official pricing page