Strengths
- In 4-bit form (INT4) it runs on a single H100 server GPU.
- Extremely cheap through providers, from $0.08 input and $0.24 output per million tokens.
- Meta tested it with a window of up to 10 million tokens, although providers offer far less.
Weaknesses
- The license does not grant rights to the multimodal Llama 4 models to people living in the EU or companies headquartered there. An EU company may not run it or build on it itself. End users of finished products are not affected.
- Knowledge stops in August 2024, and neither Slovak nor Czech is among the officially supported languages.
- Weaker at coding than Maverick: 32.8 on LiveCodeBench in Meta's table, against 43.4.
- For local use it is outdone by the newer Muse Glimmer 30B, with an Apache 2.0 license and fresher knowledge.
Best for
- cheap classification and data extraction for companies outside the EU
- existing projects built on Llama
Not ideal for
- EU companies that want to run it
- Slovak or Czech content
- complex coding
How to use it in Europe
Meta does not offer it on its own API, but you can still download the weights, and providers listed on OpenRouter host the model. The license works the same as for Maverick: people living in the EU and companies headquartered there get no rights to it, so an EU company may not run it or build on it itself. End users of finished apps are not affected. For local use, Muse Glimmer 30B is the better choice.
Price compared with similar models
Benchmark results
| Benchmark | Score | Source |
|---|---|---|
| MMLU Pro | 74.3 | dev.meta.ai ↗ (vendor-reported), April 2025 |
| MMMU | 69.4 | dev.meta.ai ↗ (vendor-reported), April 2025 |
Benchmarks cover only a narrow slice of ability and vendors often report them at their best settings. Treat them as a rough guide.
More models from this vendor
- Muse Wonder: Meta's strongest model for long-running agents and coding, still released as a preview.
- Muse Spark 1.3: Meta's balanced model with a 1-million-token window and a fair price. The Muse Spark line also powers the Meta AI assistant.
- Muse Spark Open: Meta's large open model that a company can run on its own servers, so data never leaves its infrastructure.
- Muse Glimmer 30B: Meta's small open model under Apache 2.0 that runs on a single graphics card, even offline.
- Llama 4 Maverick: The larger model in the newest Llama generation, from April 2025. Cheap through providers, but its license excludes people and companies in the EU.
Similar models from other vendors
- Gemma 4 31B (Google): Google's largest open model, which you can run on your own computer or server when data must not leave the company.
- Gemma 4 26B A4B (Google): Google's faster open model for your own server. Close to the larger Gemma 4 31B in quality, but quicker to respond.
- Ministral 3 14B (Mistral AI): Mistral's small open model under Apache 2.0 that runs on a single GPU, even without an internet connection.
- Nemotron 3.5 Bolt 74B-A9B (NVIDIA): NVIDIA's open model for AI agents on your own powerful PC, small enough to fit in 24 GB of graphics card memory.
- GPT-6 Luna (OpenAI): OpenAI's cheapest model, made for high volumes of simple tasks, sorting text and low-cost chatbots.
- gpt-oss-20b (OpenAI): OpenAI's smaller open model that runs on a computer with 16 GB of memory, even without internet access.