Granite 4.2 8B
Intelligence
11.81
#320 of 520 models
Output speed
87.2 tok/s
Measured generation rate
Cost / index task
$0.024
Measured evaluation workload
End-to-end latency
30ms
First answer 24ms
01 / Position
A model in the landscape
Compare this model with peers across intelligence, speed, cost, and individual evaluations.
02 / Evidence
Benchmark results
20 distinct evaluations grouped by topic. Explore cross-model charts or switch to the full results list.
Domain capability
03 / Operation
Speed, tokens & price
Prices are per million tokens; latency and task costs describe the captured evaluation rather than a universal provider guarantee.
Token pricing
- Input / 1M
- $0.060
- Output / 1M
- $0.250
- Cache read / 1M
- $0.015
- Cache write / 1M
- —
- Cache-hit discount
- 75%
Response & usage
- Output speed
- 87.2 tok/s
- First chunk
- 1ms
- First answer
- 24ms
- Output / task
- 33.4k
- Reasoning / task
- 21.8k
Evaluation workload
- Canonical input
- 1.19B
- Canonical output
- 125.7M
- Answer / task
- 11.6k
- Index time / task
- 6.3m
- Run cost
- $53.072
By prompt type
| Prompt | Output speed | First chunk | First answer | End to end |
|---|---|---|---|---|
| Medium | 76 tok/s | 0ms | 27ms | 33ms |
| Long | 87.2 tok/s | 1ms | 24ms | 30ms |
| Hundred K | 66.2 tok/s | 9ms | 39ms | 47ms |
| Medium Parallel | 66.4 tok/s | 1ms | 31ms | 38ms |
04 / Access
License & availability
Recorded license
Apache 2.0
The publisher-linked license text defines permitted use and redistribution. Review it before using the model.
- Weight access
- Open weights
- Classification
- Permissive
- Reasoning
- Yes
Routing and providers
- OpenRouter ID
- ibm-granite/granite-4.2-8b
- Tokenizer
- Other
- Instruct type
- —
- Captured endpoints
- 2
| Provider | Quantization | Context | tok/s |
|---|---|---|---|
| DeepInfra | bf16 | 131.1k | — |
| CoreWeave | bf16 | 131.1k | — |
05 / Audit
Underlying fields
Inspect the captured model and routing records. Artificial Analysis destination links are omitted.