will it local

Best local LLMs for the NVIDIA GeForce RTX 3080 10GB

Speed

computed band 68.0-110.0 tok/s for 8B Q4_K_M (roofline, 760.3 GB/s VRAM); community: 80.6 tok/s Meta Llama 3.1 8B Instruct Q4_K_M (https://www.localscore.ai/accelerator/132); community: 94.4 (best single run) tok/s Meta Llama 3.1 8B Instruct Q4_K_M (https://www.localscore.ai/result/3846); community: 39.3 tok/s Qwen2.5 14B Instruct Q4_K_M (https://www.localscore.ai/accelerator/132)

Effective decode window: 0.46–0.76 of 760.3 GB/s nominal → ~350–578 GB/s effective (llama.cpp decode, Q4_K_M basis; arch window, community-calibrated)

What fits (computed)

→ 14b-q4 tight 10.5 of 10 GB usable — barely over; real with q8 KV cache (halves KV) or shorter context

→ 3b-q4 full 2.8 GB needed of 10 GB usable — headroom for context

→ 4b-q4 full 3.7 GB needed of 10 GB usable — headroom for context

→ 8b-q4 full 6.1 GB needed of 10 GB usable — headroom for context

Cited community benches (3 rows)

→ 80.6 tok/s · Meta Llama 3.1 8B Instruct Q4_K_M LocalScore median · source

→ 94.4 (best single run) tok/s · Meta Llama 3.1 8B Instruct Q4_K_M LocalScore single result · source

→ 39.3 tok/s · Qwen2.5 14B Instruct Q4_K_M LocalScore median (fits, tight) · source

Runs fully in memory (machine alternatives)

Nominal bandwidth: 760.3 GB/s — see the effective decode window above

→ qwen3:0.6b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 0.6B · 0.6GB · needs ~1GB (4k ctx)

→ qwen3:1.7b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 1.7B · 1.4GB · needs ~1.8GB (4k ctx)

→ qwen3:4b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 4B · 2.6GB · needs ~3.2GB (4k ctx)

→ llama3.2:1b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 1B · 1.3GB · needs ~1.4GB (4k ctx)

→ llama3.2:3b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 3B · 2.0GB · needs ~2.4GB (4k ctx)

→ gemma3:1b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 1B · 1.0GB · needs ~1.1GB (4k ctx)

→ phi4:mini on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) mini 3.8B · 2.5GB · needs ~3GB (4k ctx)

→ moondream:2b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 2B · 1.7GB · needs ~3.7GB (4k ctx)

→ whisper:tiny on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) tiny · 0.1GB · needs ~2.1GB (4k ctx)

→ whisper on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) base · 0.3GB · needs ~2.3GB (4k ctx)

→ whisper:small on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) small · 0.9GB · needs ~2.9GB (4k ctx)

→ whisper:medium on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) medium · 3.1GB · needs ~5.1GB (4k ctx)

→ kokoro:82m on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 82M · 0.3GB · needs ~2.3GB (4k ctx)

→ gemma3:4b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 4B · 3.3GB · needs ~3.8GB (4k ctx)

→ deepseek-r1:7b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) distill 7B · 4.7GB · needs ~4.9GB (4k ctx)

→ mistral:7b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 7B · 4.1GB · needs ~4.6GB (4k ctx)

→ llava:7b on Raspberry Pi 5 + AI Kit (Hailo-8L) ($110) 7B · 4.7GB · needs ~6.7GB (4k ctx)

→ qwen3:8b on Tesla P100 (16GB, used) ($135) 8B · 5.2GB · needs ~5.8GB (4k ctx)

→ qwen2.5vl:7b on Tesla P100 (16GB, used) ($135) 7B · 5.6GB · needs ~5.8GB (4k ctx)

→ llama3.1:8b on Tesla P100 (16GB, used) ($135) 8B · 4.9GB · needs ~5.4GB (4k ctx)

→ deepseek-r1:8b on Tesla P100 (16GB, used) ($135) distill 8B · 4.9GB · needs ~5.4GB (4k ctx)

→ whisper:large-v3 on Tesla P100 (16GB, used) ($135) large-v3 · 6.2GB · needs ~8.2GB (4k ctx)

→ mistral-nemo:12b on Tesla P100 (16GB, used) ($135) 12B · 7.1GB · needs ~7.7GB (4k ctx)

→ sdxl on Tesla P100 (16GB, used) ($135) SDXL base · 6.9GB · needs ~8.9GB (4k ctx)

→ gemma3:12b on Tesla P100 (16GB, used) ($135) 12B · 8.1GB · needs ~9.6GB (4k ctx)

→ qwen3:14b on Tesla P100 (16GB, used) ($135) 14B · 9.3GB · needs ~9.9GB (4k ctx)

→ deepseek-r1:14b on Tesla P100 (16GB, used) ($135) distill 14B · 9.0GB · needs ~9.8GB (4k ctx)

→ phi4:14b on Tesla P100 (16GB, used) ($135) 14B · 9.1GB · needs ~9.9GB (4k ctx)

Borderline — runs, but offloads

→ llava:13b 13B — 11.1 GB vs 10 GB usable — partial CPU offload, expect large speed loss

Gotchas

⚠ The used-value sweet spot (GA102, 8704 cores)

⚠ 10GB fits 14B Q4_K_M (9.0 GB need) — but barely: context/KV budget is thin, watch utilization

⚠ Disambiguation: plain "RTX 3080" = this 10GB card; the 12GB refresh (rtx-3080-12gb) is a different SKU

Where to go next