Meta Llama 3.2 Vision

Llama 3.2 Vision is Meta's September 2024 11B / 90B vision snapshot, not the live Llama multimodal lead. Rechecked 2026-08-17: llama.com now leads with Llama 4 (Maverick, Scout, Behemoth). There is no separate Llama 4 page on this site. Drop leftover latest multimodal / first Llama vision as a 2026 default / unaudited MMMU crown.

Compare GPT-5.6 Sol if you wanted a current frontier chat model, Qwen3.8 if you wanted Alibaba's live multimodal flagship, and Nano Banana if you wanted Gemini image editing rather than this 2024 VLM.

Key Features

• Previous 11B / 90B VLM: Leftover copy described image+text instruct twins. Recheck Hugging Face or llama.com for cards that still download.
• Not the 2026 multimodal lead: Llama 4 is the live llama.com headline.
• License is community, not MIT: Recheck the Llama 3.2 Community License before you call it fully commercial.
• No invented bench: Do not paste leftover MMMU / ChartQA / DocVQA crowns unless you re-run a first-party card.

Use Cases

• People with a 2024 Llama 3.2 Vision bookmark who need the catalog corrected.
• People comparing leftover 11B/90B vision to Llama 4.
• People who were about to call it Meta's current multimodal.

Limitation: We did not re-download the 2024 vision cards or invent a 2026 per-token price.

Pricing

First-party pages on 2026-08-17.

If a leftover blog still calls 3.2 Vision Meta's latest multimodal, treat this page as the correction.

Getting Started

1. Open llama.com before you download 11B / 90B vision weights.
2. Recheck Llama 4 if you wanted the live Meta multimodal path.
3. Recheck Qwen3.8 if you wanted another 2026 multimodal flagship.
4. Do not paste "first Llama vision" into a 2026 deck.

Frequently Asked Questions

Is 3.2 Vision still Meta's multimodal flagship?

No. llama.com now leads with Llama 4.

Is it fully open source?

Recheck the Llama 3.2 Community License. Do not assume MIT commercial.

Same as Llama 4 Maverick?

No. Maverick is a current Llama 4 row. This slug is the 2024 vision snapshot.

Alternatives

• GPT-5.6 Sol: Live OpenAI frontier chat model.
• Qwen3.8-2.4T-A95B: Live Alibaba multimodal flagship.
• Nano Banana: Gemini official image surface.

Tips

1. Call 3.2 Vision a September 2024 snapshot.
2. Point new Meta vision work at Llama 4.
3. Recheck the community license before you invent a commercial claim.

Conclusion

Llama 3.2 Vision is a 2024 Meta VLM snapshot, not the 2026 Llama multimodal default. Start at llama.com, then decide whether Llama 4 already covers the model you need.