Granite 4.1 8B API Pricing
IBM · released 2026-04-29
Pricing
< 0.01¢ / 1K
0.010¢ / 1K
< 0.01¢ / 1K
What it actually costs
| Workload | Tokens | Cost |
|---|---|---|
| Short chat reply | 500 in / 300 out | 0.0055¢ |
| RAG query | 4,000 in / 500 out | 0.025¢ |
| Long document summary | 50,000 in / 1,000 out | 0.26¢ |
| Code review of a 2k-line file | 30,000 in / 2,000 out | 0.17¢ |
| 1,000 chat replies/day for 30 days | 30,000 × (500 in / 300 out) | $1.65 |
Cost = (input tokens ÷ 1,000,000 × $0.05) + (output tokens ÷ 1,000,000 × $0.10).
Benchmarks
#407 of 533
#178 of 197
Similar intelligence, lower price
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Llama 3.1 Instruct 8B | 6.9 | $0.028 |
| Gemma 4 E4B (Reasoning) | 8.9 | $0.04 |
| Sarvam 30B (high) | 6.6 | $0.047 |
| Granite 4.2 3B | 9.1 | $0.052 |
| Nova Micro | 5.9 | $0.061 |
Leaderboard neighbours
| Model | Intelligence | Blended $/1M |
|---|---|---|
| DeepSeek-V2.5 | 6.6 | — |
| DeepSeek-V2.5 (Dec '24) | 6.6 | — |
| Gemini 2.0 Flash Thinking Experimental (Dec '24) | 6.6 | — |
| Granite 4.1 8B | 6.6 | $0.063 |
| Llama 3.1 Instruct 70B | 6.6 | $0.56 |
| Qwen3 30B A3B (Non-reasoning) | 6.6 | $0.35 |
| Qwen3 4B (Non-reasoning) | 6.6 | — |
More from IBM
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Granite 4.2 30B | 14.8 | $0.282 |
| Granite 4.2 8B | 11.1 | $0.107 |
| Granite 4.2 3B | 9.1 | $0.052 |
| Granite 4.1 30B | 7.4 | — |
| Granite 4.0 H Small | 6.0 | $0.107 |
| Granite 4.1 3B | 5.9 | — |
| Granite 4.0 H 1B | 5.2 | — |
| Granite 4.0 Micro | 5.1 | — |
Count tokens & estimate cost for Granite 4.1 8B See Granite 4.1 8B on the leaderboard
Source: Artificial Analysis · last synced 2026-09-22
FAQ
Who develops Granite 4.1 8B?
Granite 4.1 8B is developed by IBM, released 2026-04-29.
How much does Granite 4.1 8B cost per million tokens?
Granite 4.1 8B costs $0.05 per 1M input tokens and $0.10 per 1M output tokens ($0.063 blended), per Artificial Analysis's official pricing data.
How does Granite 4.1 8B rank on benchmarks?
Granite 4.1 8B scores 6.6 on the Artificial Analysis Intelligence Index, ranking #407 of 533 models we track.
Is there a cheaper model with similar intelligence to Granite 4.1 8B?
Yes — Llama 3.1 Instruct 8B scores similar intelligence (6.9) at $0.028/1M blended, versus $0.063/1M for Granite 4.1 8B.
Where does this data come from?
Pricing and benchmark data for Granite 4.1 8B comes from Artificial Analysis (artificialanalysis.ai), synced daily. Last synced 2026-09-22.