Model Library
IBMReleased September 22, 2025Granite 4.0 H Small
IBM · Released September 22, 2025
Intelligence score
5
At a glance
- Intelligence
- 5
- Speed
- 383.7tokens/sec
- Latency
- 10.48stime to first token
- Input price
- $0.06per 1M tokens
- Output price
- $0.25per 1M tokens
- Context window
- 130Ktokens
- Parameters
- 32billions
Summary
Comparison summary
Granite 4.0 H Small scores 5 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 6). Granite 4.0 H Small generates output at 383.7 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s). Granite 4.0 H Small costs $0.06 per 1M input tokens (very competitive, median: $0.15) and $0.25 per 1M output tokens (better than average, median: $0.32), based on the median across providers serving the model.
Specifications
Technical specifications
| Model name | Granite 4.0 H Small |
|---|---|
| Provider | IBM |
| Release date | September 22, 2025 |
| Intelligence score | 5 |
| Speed | 383.7 tokens/sec |
| Latency (TTFT) | 10.48s |
| Input price | $0.06 / 1M tokens |
| Output price | $0.25 / 1M tokens |
| Context window | 130,000 tokens |
| Parameters | 32B |
| Reasoning | No |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | Apache 2 |
| API availability | — |
| Source | Model Intelligence legacy profile: granite-4-0-h-small |
Pricing
Token prices
Price per 1M tokens
Granite 4.0 H Small, as listed in the catalog record
- Input$0.06
- Output$0.25
| Token type | Price per 1M |
|---|---|
| Input | $0.06 |
| Output | $0.25 |
Model Intelligence legacy profile: granite-4-0-h-small
FAQ · 20
Frequently asked questions
Granite 4.0 H Small was released on September 22, 2025.
Granite 4.0 H Small was created by IBM.
Granite 4.0 H Small scores 5 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 6).
Granite 4.0 H Small generates output at 383.7 tokens per second (based on the median across providers serving the model), which is well above average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s).
Granite 4.0 H Small has a time to first token (TTFT) of 10.48s (based on the median across providers serving the model), which is at the higher end compared to other open weight non-reasoning models of similar size (median: 1.54s).
Granite 4.0 H Small costs $0.06 per 1M input tokens (very competitive, median: $0.15) and $0.25 per 1M output tokens (better than average, median: $0.32), based on the median across providers serving the model.
Granite 4.0 H Small costs $0.06 per 1M input tokens and $0.25 per 1M output tokens (based on the median across providers serving the model). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.08 per 1M tokens. Pricing may vary by provider.
No, Granite 4.0 H Small is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Granite 4.0 H Small supports text input.
Granite 4.0 H Small supports text output.
No, Granite 4.0 H Small does not support image input. It can only process text.
No, Granite 4.0 H Small is not multimodal. It only supports text input.
Granite 4.0 H Small has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Granite 4.0 H Small is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Granite 4.0 H Small has 32 billion parameters (9 billion active).
Granite 4.0 H Small is a Mixture of Experts (MoE) model with 32 billion total parameters, but only 9 billion active parameters are used during inference.
Granite 4.0 H Small is released under the Apache 2.0 license. This license allows commercial use.
Granite 4.0 H Small achieves a score of 5 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Yes, Granite 4.0 H Small is available via API through 1 provider.
Granite 4.0 H Small is available through 1 API provider.