Granite 4.1 8B
AI model by IBM. Real-time pricing and benchmark data.
Benchmarks
Editorial Analysis
Profile. Granite 4.1 8B is a budget-tier option from IBM at $0.10/M output tokens. Despite the low price, it scores competitively on Coding Index (9.5) and Intelligence Index (6.6) — making it a candidate for high-volume production workloads where unit cost dominates.
Latency. On latency, Granite 4.1 8B reports 126 tok/s output speed and 740ms time-to-first-token. Output speed is in the normal range — most chat and completion applications will be token-bound, not streaming-bound.
Cost model. For a workload of 1M input tokens + 100K output tokens, Granite 4.1 8B costs $0.06 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.
Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-10-08.
Compare with similar models
| Model | Input | Output | Speed |
|---|---|---|---|
Granite 4.1 8BCurrent | $0.05 | $0.10 | 126 tok/s |
| Gemma 4 E4B (Non-reasoning) | $0.02 | $0.10 | 45 tok/s |
| Ministral 3 3B | $0.10 | $0.10 | 214 tok/s |
| Gemma 4 E4B (Reasoning) | $0.02 | $0.10 | 44 tok/s |
| Sarvam 30B (high) | $0.03 | $0.11 | — |
| HyperNova 60B 2605 (high, based on gpt-oss-120b) | $0.04 | $0.14 | 353 tok/s |