will it local
68 GB/s LPDDR5 shared; 8GB total — usable ≈ 6GB after JetPack Linux; 8B Q4 does NOT fit, 3-4B tight Super-mode (JP6) boosts clocks ~50%: 3-4B Q4 ~8-11 tok/s community rows 1024 CUDA cores of Ampere — tensor cores help prefill, bandwidth walls decode
Shared-memory device — speed scales with RAM bandwidth (68.0 GB/s shared RAM), usable for inference ≈ 6.0 GB of 8 GB system RAM
→explore the full catalogupgrade paths with live prices · what fits on each machinecomputed band 3.3-8.6 tok/s for 8B Q4_K_M (roofline, 68.0 GB/s shared RAM); community: 8-11 tok/s 3B Q4_K_M (community Metal/Vulkan rows (gotcha-cited, 0905))
Effective decode window: 0.25–0.65 of 68 GB/s nominal → ~17–44 GB/s effective (llama.cpp decode, Q4_K_M basis; arch window, community-calibrated; nominal is LPDDR/JEDEC channel arithmetic (shared memory))
→ 13b-q4 no 14.6 GB vs 6 GB usable — does not fit
→ 3b-q4 full 2.8 GB needed of 6 GB usable — headroom for context
→ 4b-q4 full 3.7 GB needed of 6 GB usable — headroom for context
→ 8b-q4 tight 6.1 of 6 GB usable — barely over; real with q8 KV cache (halves KV) or shorter context
→ 8-11 tok/s · 3B Q4_K_M · community Metal/Vulkan rows (gotcha-cited, 0905)
Nominal bandwidth: 68.0 GB/s shared memory — see the effective decode window above
→ qwen3:0.6b on Tesla M60 (16GB, used) ($35) 0.6B · 0.6GB · needs ~1GB (4k ctx)
→ qwen3:1.7b on Tesla M60 (16GB, used) ($35) 1.7B · 1.4GB · needs ~1.8GB (4k ctx)
→ qwen3:4b on Tesla M60 (16GB, used) ($35) 4B · 2.6GB · needs ~3.2GB (4k ctx)
→ llama3.2:1b on Tesla M60 (16GB, used) ($35) 1B · 1.3GB · needs ~1.4GB (4k ctx)
→ llama3.2:3b on Tesla M60 (16GB, used) ($35) 3B · 2.0GB · needs ~2.4GB (4k ctx)
→ gemma3:1b on Tesla M60 (16GB, used) ($35) 1B · 1.0GB · needs ~1.1GB (4k ctx)
→ phi4:mini on Tesla M60 (16GB, used) ($35) mini 3.8B · 2.5GB · needs ~3GB (4k ctx)
→ moondream:2b on Tesla M60 (16GB, used) ($35) 2B · 1.7GB · needs ~3.7GB (4k ctx)
→ whisper:tiny on Tesla M60 (16GB, used) ($35) tiny · 0.1GB · needs ~2.1GB (4k ctx)
→ whisper on Tesla M60 (16GB, used) ($35) base · 0.3GB · needs ~2.3GB (4k ctx)
→ whisper:small on Tesla M60 (16GB, used) ($35) small · 0.9GB · needs ~2.9GB (4k ctx)
→ whisper:medium on Tesla M60 (16GB, used) ($35) medium · 3.1GB · needs ~5.1GB (4k ctx)
→ kokoro:82m on Tesla M60 (16GB, used) ($35) 82M · 0.3GB · needs ~2.3GB (4k ctx)
→ gemma3:4b on Tesla M60 (16GB, used) ($35) 4B · 3.3GB · needs ~3.8GB (4k ctx)
→ deepseek-r1:7b on Tesla M60 (16GB, used) ($35) distill 7B · 4.7GB · needs ~4.9GB (4k ctx)
→ mistral:7b on Tesla M60 (16GB, used) ($35) 7B · 4.1GB · needs ~4.6GB (4k ctx)
→ qwen3:8b on Tesla M60 (16GB, used) ($35) 8B · 5.2GB · needs ~5.8GB (4k ctx)
→ qwen2.5vl:7b on Tesla M60 (16GB, used) ($35) 7B · 5.6GB · needs ~5.8GB (4k ctx)
→ llama3.1:8b on Tesla M60 (16GB, used) ($35) 8B · 4.9GB · needs ~5.4GB (4k ctx)
→ deepseek-r1:8b on Tesla M60 (16GB, used) ($35) distill 8B · 4.9GB · needs ~5.4GB (4k ctx)
→ llava:7b 7B — 6.7 GB vs 6 GB usable — partial CPU offload, expect large speed loss
⚠ 68 GB/s LPDDR5 shared; 8GB total — usable ≈ 6GB after JetPack Linux; 8B Q4 does NOT fit, 3-4B tight
⚠ Super-mode (JP6) boosts clocks ~50%: 3-4B Q4 ~8-11 tok/s community rows
⚠ 1024 CUDA cores of Ampere — tensor cores help prefill, bandwidth walls decode
—