Maverick and Goose!
Llama 4 Maverick:
- 17B active parameters, 128 experts, 400B total.
- 1M token context window.
- Not single-GPU; runs on one H100 DGX host or can be distributed for greater efficiency.
- Outperforms GPT-4o and Gemini 2.0 Flash on coding, reasoning, and multilingual tests at a competitive cost.
- Maintains strong image understanding and grounded reasoning ability.
On paper and from what Zuck was saying, sounds amazing!
I still have to give a test run
Please Login to reply.
No replies yet.
