IBM
Granite 4.1 8B
Tool-capable chat model for instruction following and agentic application workflows
Model facts
- Context window
- 131K tokens
- Maximum output
- 118K tokens
- Cheapest paid input
- $0.05 per 1M tokens
- Cheapest paid output
- $0.1 per 1M tokens
- Open weights
- yes
- Providers
- 5
Capabilities and modalities
tool calling, structured output, temperature control, open weights, text.
Local hardware estimate
At 4K context: Q4_K_M 6.4 GiB, Q8_0 10.5 GiB, F16 18.1 GiB working memory.
- Parameters
- 8.8 billion
- Architecture
- granite
- Model license
- apache-2.0
Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.
Artificial Analysis benchmarks
Compare this model on the full LLM benchmark leaderboard.
| Configuration | AAI | Coding | Math | Output tokens/s |
|---|---|---|---|---|
| default | 6.4 | 9.5 | not measured | 105.7 |
Observed price history
- 2026-08-24: $0.05 input / $0.1 output per 1M tokens
API providers
- OpenRouter — model id ibm-granite/granite-4.1-8b; $0.05 input / $0.1 output per 1M tokens; provider documentation
- Kilo Gateway (IBM) — model id ibm-granite/granite-4.1-8b; $0.05 input / $0.1 output per 1M tokens; provider documentation
- NanoGPT — model id ibm-granite/granite-4.1-8b; $0.05 input / $0.1 output per 1M tokens; provider documentation
- Weights & Biases — model id ibm-granite/granite-4.1-8b; $0.05 input / $0.1 output per 1M tokens; provider documentation
- Arena — model id granite-4.1-8b; not published input / not published output per 1M tokens; provider documentation