The closest alternative to Llama 3 8B Instruct is Gemma 3 4B — comparable intelligence (6.3 vs 6.4); for a cheaper option, Nemotron 3 Ultra (free) runs Free/M vs $0.140/M; for more speed, gpt-oss-120b (free) hits 554 t/s; and the best open-weight substitute is MiniMax M3.
Gemma 3 4B is the closest alternative to Llama 3 8B Instruct, with an Intelligence Index of 6.3 (Llama 3 8B Instruct scores 6.4) at $0.050 per million input tokens.
Yes — Nemotron 3 Ultra (free) costs Free per million input tokens versus $0.140 for Llama 3 8B Instruct, while staying within ~8 points of its Intelligence Index.
MiniMax M3 is the strongest open-weight alternative to Llama 3 8B Instruct — downloadable from Hugging Face and self-hostable, scoring 54.7 on intelligence.