Granite 4.1 30B vs Granite 4.2 30B
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models
Granite 4.1 30B4.1
Granite 4.2 30B11.2
Terminal-Bench 4.0 Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite
Granite 4.1 30B—
Granite 4.2 30B0
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
Granite 4.1 30B—
Granite 4.2 30B1.5
Arena Text · creative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool
Granite 4.1 30B—
Granite 4.2 30B1,266
| Benchmark | Granite 4.1 30B | Granite 4.2 30B |
|---|---|---|
| Humanity's Last Exam | 4.1 | 11.2 |
| Terminal-Bench 4.0 | — | 0 |
| AutomationBench-AA | — | 1.5 |
| Arena Text · creative writing | — | 1,266 |
Specs
| Spec | Granite 4.1 30B | Granite 4.2 30B |
|---|---|---|
| Input, per 1M | $0.10 | $0.16 |
| Output, per 1M | $0.25 | $0.65 |
| Cost per task? | — | $0.072 |
| Speed, tok/s | — | 48 |
| Context | 131K | 128K |
| Weights | Open | Open |
| Released | Apr 2026 | Aug 2026 |
| Memory (4-bit) | 17 GB | 18 GB |
| Licence | Apache-2.0 | Apache-2.0 |
| Commercial use | Allowed | Allowed |
| See images | — | — |
| Read PDFs | ? | ? |
| Hear audio | — | — |
| Watch video | — | — |
| Call tools | ✓ | ✓ |
| Force a tool call | ? | ? |
| Return strict JSON | Partial | Partial |
| Search the web | ? | ? |
| Run code | ? | ? |
| Use a computer | ? | ? |
What breaks when you switch
Granite 4.1 30B → Granite 4.2 30B
Context: 131K → 128K
Memory (4-bit): 17 GB → 18 GB
Granite 4.2 30B → Granite 4.1 30B
Nothing known.