Granite 4.0 1B vs Granite 4.2 3B
Different efforts: non-thinking · thinking
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models
Granite 4.0 1B4.8
Granite 4.2 3B6.6
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 4.0 1B—
Granite 4.2 3B0
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
Granite 4.0 1B—
Granite 4.2 3B0.4
Arena Text · creative writing Arena Text · creative writing: Human pairwise preferences in the stated text vote pool
Granite 4.0 1B—
Granite 4.2 3B1,206
| Benchmark | Granite 4.0 1B | Granite 4.2 3B |
|---|---|---|
| Humanity's Last Exam | 4.8 | 6.6 |
| Terminal-Bench 4.0 | — | 0 |
| AutomationBench-AA | — | 0.4 |
| Arena Text · creative writing | — | 1,206 |
Specs
| Spec | Granite 4.0 1B | Granite 4.2 3B |
|---|---|---|
| Input, per 1M | — | $0.03 |
| Output, per 1M | — | $0.12 |
| Cost per task? | — | $0.006 |
| Speed, tok/s | — | 56 |
| Context | 128K | 128K |
| Weights | Open | Open |
| Released | Oct 2025 | Aug 2026 |
| Memory (4-bit) | 0.8 GB | 1.8 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 4.0 1B → Granite 4.2 3B
Memory (4-bit): 0.8 GB → 1.8 GB
Granite 4.2 3B → Granite 4.0 1B
Nothing known.