What hardware do you need to run Granite 4.2 8B?
Granite 4.2 8B at Q4_K_M is 5.3 GB of weights, with a context ceiling of 128k tokens. Not yet rated: released after our last ratings pass. IBM's small model, aimed at enterprises that care about provenance.
How good is it, really?
On the Artificial Analysis Intelligence Index v4.3 it scores 12, which puts it in the Below every hosted tier band. Fine for simple, well-specified tasks. Noticeably less capable than anything the big labs sell today. Score source. See the whole table.
- Summarising — not rated
- Translation — not rated
- Everyday coding — not rated
- Reasoning & maths — not rated
- Agentic work — not rated
What it costs either way
Renting the same model costs $0.06 per million input tokens and $0.25 per million output (OpenRouter, cheapest active endpoint, checked 2026-09-09). Buying a machine only beats that if you use it hard enough, for long enough, that the hardware price divides down below the rental bill.
Machines that run it
| Machine | Price | Speed at 32k | Pay-back | |
|---|---|---|---|---|
| Mac mini M6, 24GB | $1,099 | 12 tok/s estimated | Pays back in 108 years | Run the numbers |
| Strix Halo Framework Desktop, 32GB | $1,269 | 18 tok/s estimated | Pays back in 139 years | Run the numbers |
| MacBook Pro M5 (14-inch), 32GB | $2,399 | 11 tok/s estimated | Pays back in 243 years | Run the numbers |
| Mac Studio M5 Max, 36GB | $2,499 | 32 tok/s estimated | Pays back in 234 years | Run the numbers |
| DGX Spark GB10 Grace Blackwell, 128GB | $4,699 | 19 tok/s estimated | Pays back in 528 years | Run the numbers |
One machine per family, cheapest first. Speeds are measured where a public benchmark exists and estimated from memory bandwidth otherwise; the calculator says which for any configuration.
The specifics
- Parameters
- 8B
- Quantisation
- Q4_K_M
- Weights on disk
- 5.3 GB
- KV cache
- 5.4 GB at 32k context — Plain grouped-query attention on all 40 layers, so the context costs real memory. head_dim is absent from the config and derived as hidden_size ÷ heads.
- Maximum context
- 128k tokens (128k)
- Licence
- Apache 2.0
- Sources
- source 1