IBMGranite 4.2 30B
Best configuration ranks #269 of 610 on the BenchLeader Index at 51.3 ±2.5. Last measured 2 Sept 2026. Released 25 Aug 2026.
- Blended price
- $0.283/M
- $0.160 in · $0.650 out
- Output speed
- 76 tok/s
- First answer
- 27 s
- Context
- 131k
- Overall index51
- Reasoning52
- Coding53
- Knowledge58
- Human preference51
- Long context50
- Composite48
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | default | Source |
|---|---|---|
| LMArena Hard Prompts | 1363#180 | LMArena |
| GPQA Diamond (AA)not in index | 64.4%#331 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 11.2%#234 | Artificial Analysis |
Coding
| Benchmark | default | Source |
|---|---|---|
| LMArena Coding | 1396#175 | LMArena |
| SciCode (AA)not in index | 37.9%#127 | Artificial Analysis |
Knowledge
| Benchmark | default | Source |
|---|---|---|
| AA-Omniscience | -12.9#155 | Artificial Analysis |
Human preference
| Benchmark | default | Source |
|---|---|---|
| LMArena Text | 1342#189 | LMArena |
Long context
| Benchmark | default | Source |
|---|---|---|
| AA-LCR | 49.0%#259 | Artificial Analysis |
Composite
| Benchmark | default | Source |
|---|---|---|
| AA Intelligence Index | 14.8#226 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0003 | – | 30.9 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0023 | – | 34.9 s |
| Code edit | 6,000 / 1,500 | $0.0019 | – | 46.7 s |
| Agentic coding session | 60,000 / 4,000 | $0.012 | – | 1.3 min |
| Structured extraction | 2,000 / 200 | $0.0005 | – | 29.6 s |
See also
Data as of 9 Sept 2026. Compare with another model.