Home › Best › best LLM for 8GB VRAM The best open LLMs you can run on 8 GB of VRAM Open LLMs calculated to need no more than 8 GB of VRAM at the stated default quant and context — the size of an RTX 3060/4060 or an 8 GB laptop GPU. Ranked by popularity and recognition within that calculated fit set, with the $0 compute route on hardware you already own. Verify the selected quant, context, and runner against your exact machine before downloading.
How this is ranked: Transparent calculated filter: 'best' = 'estimated to fit your 8 GB card at the stated quant and context.' The shared engine derives VRAM from parameters, quantization and context; it is not a hardware test or quality opinion. Within the set we order by popularity/recognition and never claim a #1 is 'smartest.'
1. Llama 3.1 8B Instruct — Meta, 8B · ~8.0 GB VRAM · $0.07/1M API · commercial OK2. Qwen2.5 7B Instruct — Alibaba, 7B · ~7.0 GB VRAM · $0.15/1M API · commercial OK3. Qwen3 0.6B — Qwen, 800M · ~3.0 GB VRAM · $0.11/1M API est. · commercial OK4. Mistral 7B Instruct v0.3 — Mistral AI, 7.2B · ~8.0 GB VRAM · $0.20/1M API · commercial OK5. Qwen2.5-Coder 7B Instruct — Alibaba, 7B · ~7.0 GB VRAM · $0.16/1M API est. · commercial OK6. Qwen3 4B — Qwen, 4B · ~5.0 GB VRAM · $0.13/1M API est. · commercial OK7. Qwen2.5 3B Instruct — Qwen, 3.1B · ~4.0 GB VRAM · $0.12/1M API est. · non-commercial8. Llama 3.2 3B Instruct — Meta, 3B · ~5.0 GB VRAM · $0.19/1M API · commercial OK9. Qwen2.5 1.5B Instruct — Qwen, 1.5B · ~3.0 GB VRAM · $0.11/1M API est. · commercial OK10. gemma 3 270m — google, 300M · ~2.0 GB VRAM · $0.10/1M API est. · commercial OK11. Qwen3 1.7B — Qwen, 2B · ~5.0 GB VRAM · $0.12/1M API est. · commercial OK12. BGE-M3 — BAAI, 567M · ~3.0 GB VRAM · $0.10/1M API est. · commercial OK13. Qwen3 4B Instruct 2507 — Qwen, 4B · ~5.0 GB VRAM · $0.13/1M API est. · commercial OK14. Qwen2.5 0.5B Instruct — Qwen, 500M · ~2.0 GB VRAM · $0.10/1M API est. · commercial OK15. Qwen2 1.5B Instruct — Qwen, 1.5B · ~3.0 GB VRAM · $0.11/1M API est. · commercial OK16. Llama 3.2 1B Instruct — Meta, 1.2B · ~3.0 GB VRAM · $0.11/1M API · commercial OK17. Llama 3.2 1B — meta-llama, 1.2B · ~3.0 GB VRAM · $0.11/1M API est. · commercial OK18. Qwen2.5 0.5B — Qwen, 500M · ~2.0 GB VRAM · $0.10/1M API est. · commercial OK19. Phi-3.5-mini Instruct — Microsoft, 3.8B · ~5.0 GB VRAM · $0.13/1M API est. · commercial OK20. Qwen2-VL 7B Instruct — Alibaba, 8B · ~7.0 GB VRAM · $0.16/1M API est. · commercial OK21. gemma 3 1b it — google, 1B · ~3.0 GB VRAM · $0.11/1M API est. · commercial OK22. TinyLlama 1.1B Chat v1.0 — TinyLlama, 1.1B · ~2.0 GB VRAM · $0.11/1M API est. · commercial OK23. Rio 3.0 Open Mini — prefeitura-rio, 4B · ~5.0 GB VRAM · $0.13/1M API est. · commercial OK24. gpt2 large — openai-community, 800M · ~2.0 GB VRAM · $0.11/1M API est. · commercial OK25. OpenELM 1 1B Instruct — apple, 1.1B · ~3.0 GB VRAM · $0.11/1M API est. · non-commercial26. Hermes 3 — Llama 3.1 8B — Nous Research, 8B · ~8.0 GB VRAM · $0.16/1M API est. · commercial OK27. PowerMoE 3b — ibm-research, 3.4B · ~4.0 GB VRAM · $0.13/1M API est. · commercial OK28. Mistral 7B Instruct v0.2 — mistralai, 7.2B · ~8.0 GB VRAM · $0.16/1M API est. · commercial OK29. Phi 4 mini instruct — microsoft, 3.8B · ~6.0 GB VRAM · $0.22/1M API · commercial OK30. Qwen2.5 1.5B — Qwen, 1.5B · ~3.0 GB VRAM · $0.11/1M API est. · commercial OK31. NVIDIA Nemotron 3 Nano 4B BF16 — nvidia, 4B · ~6.0 GB VRAM · $0.13/1M API est. · non-commercial32. h2ovl mississippi 800m — h2oai, 800M · ~2.0 GB VRAM · $0.11/1M API est. · commercial OK33. h2ovl mississippi 2b — h2oai, 2.2B · ~4.0 GB VRAM · $0.12/1M API est. · commercial OK34. Nomic Embed Text v1.5 — Nomic AI, 137M · ~2.0 GB VRAM · $0.10/1M API est. · commercial OK35. Llama 3.1 8B Instruct (Abliterated) — mlabonne (community), 8B · ~8.0 GB VRAM · $0.16/1M API est. · commercial OK36. Qwen2 0.5B — Qwen, 500M · ~2.0 GB VRAM · $0.10/1M API est. · commercial OK37. SmolLM 1.7B Instruct quantized.w4a16 — nm-testing, 1.8B · ~3.0 GB VRAM · $0.11/1M API est. · commercial OK38. Qwen2.5 1.5B quantized.w8a8 — RedHatAI, 1.8B · ~3.0 GB VRAM · $0.11/1M API est. · commercial OK39. LLaDA 8B Instruct — GSAI-ML, 8B · ~8.0 GB VRAM · $0.16/1M API est. · commercial OK40. Qwen2.5 Math 1.5B — Qwen, 1.5B · ~3.0 GB VRAM · $0.11/1M API est. · commercial OKShowing the top 40 of 252. See all →
Want every model that fits? All 252 models that run on 8 GB of VRAM →
More: all "best" lists · Outcome Lab · all models
Open the free LLM cost calculator → · $0 markup. © 2026 Cynosure LLC.