Quantization

For most developers in 2026, Gemma 4 31B is the best all-around open model. It ranks #3 on the LMArena leaderboard, scores 85.2% on MMLU Pro, and ships under Apache 2.0 with zero usage limits. Qwen 3.5 27B edges it on coding, and its Omni variant offers real-time speech output that no other open model matches. Llama 4 Maverick (400B MoE) wins on raw scale, but it needs datacenter hardware and Meta’s restrictive 700M MAU license. So pick Gemma 4 for the best quality-to-size ratio, Qwen 3.5 for coding-heavy work, and Llama 4 only when you need the largest open model.

Small Language Models, sub-4B parameter models built to run on local hardware, now handle most of the edge AI work that used to need the cloud. Phi-4 , Gemma 3 , and Llama 3.2-1B run offline on Raspberry Pi boards, phones, and industrial PLCs. The economics, latency, and privacy story all point the same way: edge first.

What Counts as a Small Language Model

In 2023, “small” meant under 13B parameters. Today, three tiers matter for edge work.