MetaLlama 3.2 Instruct 11B (Vision)
Best configuration ranks #579 of 610 on the BenchLeader Index at 36.4 ±6.8. Last measured 27 Mar 2025stale: no new result in six months. Released 18 Sept 2024.
- Blended price
- $0.350/M
- $0.350 in · $0.350 out
- Output speed
- 13 tok/s
- First answer
- 1.24 s
- first token 1.24 s
- Context
- 128k
- Overall index36
- Agents & tools34
- Knowledge34
- Instruction following34
- Multimodal14
- Long context31
- Composite36
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 |
|---|---|---|
| GPQA Diamond (AA)not in index | 22.1%#521 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 5.5%#346 | Artificial Analysis |
Agents & tools
| Benchmark | default | Source |
|---|---|---|
| Terminal-Bench Hard | 0.8%#332 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 14.6%#349 | Artificial Analysis |
Knowledge
| Benchmark | default | Source |
|---|---|---|
| AA-Omniscience | -62.6#405 | Artificial Analysis |
Instruction following
| Benchmark | default | Source |
|---|---|---|
| IFBench | 30.4%#343 | Artificial Analysis |
Multimodal
| Benchmark | default | Source |
|---|---|---|
| MMMU-Pro | 29.3%#240 | Artificial Analysis |
| VISTA | 20.5%#56 | Scale AI SEAL |
Long context
| Benchmark | default | Source |
|---|---|---|
| AA-LCR | 12.7%#388 | Artificial Analysis |
Composite
| Benchmark | default | Source |
|---|---|---|
| AA Intelligence Index | 5.4#492 | 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 | $0.0002 | – | 24.9 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0044 | – | 48.5 s |
| Code edit | 6,000 / 1,500 | $0.0026 | – | 2.0 min |
| Agentic coding session | 60,000 / 4,000 | $0.022 | – | 5.3 min |
| Structured extraction | 2,000 / 200 | $0.0008 | – | 17.0 s |
See also
Data as of 9 Sept 2026. Compare with another model.