What can NVIDIA GeForce RTX 5080 run for long context?
Build: NVIDIA GeForce RTX 5080 + — + 32 GB RAM (windows)
Memory: 16 GB VRAM + 32 GB system RAM
Runner: llama.cpp / Ollama (CUDA)
Runs comfortably101 models
Ranked by fit for long context use case + predicted speed. Click a row for VRAM breakdown.
Quant: Q4_K_MContext: 8,192VRAM: 10.8 GBHeadroom: 5.2 GBTTFT: fastollama run qwen3:8b129tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 10.8 GBHeadroom: 5.2 GBTTFT: fast
ollama run qwen3:8b129
tok/s
Estimated
Weights
4.80 GB
KV cache
4.00 GB
Activations
0.25 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~366 ms (fast)
Quant: Q8_0Context: 8,192VRAM: 8.4 GBHeadroom: 7.6 GBTTFT: fastollama run gemma4:e4b147tok/sEstimated
Quant: Q8_0Context: 8,192VRAM: 8.4 GBHeadroom: 7.6 GBTTFT: fast
ollama run gemma4:e4b147
tok/s
Estimated
Weights
4.40 GB
KV cache
2.00 GB
Activations
0.23 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~183 ms (fast)
Quant: Q8_0Context: 8,192VRAM: 8.0 GBHeadroom: 8.0 GBTTFT: fastollama run phi3.5:3.8b155tok/sEstimated
Quant: Q8_0Context: 8,192VRAM: 8.0 GBHeadroom: 8.0 GBTTFT: fast
ollama run phi3.5:3.8b155
tok/s
Estimated
Weights
4.10 GB
KV cache
1.90 GB
Activations
0.21 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~174 ms (fast)
Quant: Q4_K_MContext: 8,192VRAM: 10.4 GBHeadroom: 5.6 GBTTFT: fast129tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 10.4 GBHeadroom: 5.6 GBTTFT: fast
129
tok/s
Estimated
Weights
4.40 GB
KV cache
4.00 GB
Activations
0.23 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~366 ms (fast)
Quant: Q4_K_MContext: 8,192VRAM: 9.7 GBHeadroom: 6.3 GBTTFT: fast148tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 9.7 GBHeadroom: 6.3 GBTTFT: fast
148
tok/s
Estimated
Weights
4.20 GB
KV cache
3.50 GB
Activations
0.22 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~320 ms (fast)
Quant: Q4_K_MContext: 8,192VRAM: 9.7 GBHeadroom: 6.3 GBTTFT: fast148tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 9.7 GBHeadroom: 6.3 GBTTFT: fast
148
tok/s
Estimated
Weights
4.20 GB
KV cache
3.50 GB
Activations
0.22 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~320 ms (fast)
Quant: Q4_K_MContext: 8,192VRAM: 5.3 GBHeadroom: 10.7 GBTTFT: fast345tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 5.3 GBHeadroom: 10.7 GBTTFT: fast
345
tok/s
Estimated
Weights
1.90 GB
KV cache
1.50 GB
Activations
0.10 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~137 ms (fast)
Quant: Q4_K_MContext: 8,192VRAM: 11.1 GBHeadroom: 4.9 GBTTFT: fast129tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 11.1 GBHeadroom: 4.9 GBTTFT: fast
129
tok/s
Estimated
Weights
5.00 GB
KV cache
4.00 GB
Activations
0.26 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~366 ms (fast)
Quant: Q8_0Context: 8,192VRAM: 10.8 GBHeadroom: 5.2 GBTTFT: fastollama run qwen2.5:7b84tok/sEstimated
Quant: Q8_0Context: 8,192VRAM: 10.8 GBHeadroom: 5.2 GBTTFT: fast
ollama run qwen2.5:7b84
tok/s
Estimated
Weights
8.10 GB
KV cache
0.47 GB
Activations
0.41 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~320 ms (fast)
Quant: Q8_0Context: 8,192VRAM: 11.8 GBHeadroom: 4.2 GBTTFT: fastollama run llama3.1:8b73tok/sEstimated
Quant: Q8_0Context: 8,192VRAM: 11.8 GBHeadroom: 4.2 GBTTFT: fast
ollama run llama3.1:8b73
tok/s
Estimated
Weights
8.50 GB
KV cache
1.07 GB
Activations
0.43 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~366 ms (fast)
Quant: Q4_K_MContext: 8,192VRAM: 9.9 GBHeadroom: 6.1 GBTTFT: fast148tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 9.9 GBHeadroom: 6.1 GBTTFT: fast
148
tok/s
Estimated
Weights
4.40 GB
KV cache
3.50 GB
Activations
0.23 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~320 ms (fast)
Quant: Q4_K_MContext: 8,192VRAM: 9.4 GBHeadroom: 6.6 GBTTFT: fast148tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 9.4 GBHeadroom: 6.6 GBTTFT: fast
148
tok/s
Estimated
Weights
3.90 GB
KV cache
3.50 GB
Activations
0.20 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~320 ms (fast)
Runs with tradeoffs94 models
Tight VRAM, partial CPU offload, or context-limited.
Quant: Q4_K_MContext: 2,048VRAM: 24.5 GBHeadroom: 10.7 GBTTFT: noticeable- • Partial CPU offload: ~35% of layers run on CPU
ollama run nemotron3:nano34tok/sEstimated
Quant: Q4_K_MContext: 2,048VRAM: 24.5 GBHeadroom: 10.7 GBTTFT: noticeable
- • Partial CPU offload: ~35% of layers run on CPU
ollama run nemotron3:nano34
tok/s
Estimated
Weights
18.00 GB
KV cache
3.75 GB
Activations
0.90 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~1371 ms (noticeable)
Quant: Q4_K_MContext: 8,192VRAM: 15.7 GBHeadroom: 0.3 GBTTFT: noticeable- • Tight VRAM fit — only 0.3 GB headroom left for context growth
ollama run mistral-nemo:12b86tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 15.7 GBHeadroom: 0.3 GBTTFT: noticeable
- • Tight VRAM fit — only 0.3 GB headroom left for context growth
ollama run mistral-nemo:12b86
tok/s
Estimated
Weights
7.50 GB
KV cache
6.00 GB
Activations
0.38 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~549 ms (noticeable)
Quant: Q4_K_MContext: 2,048VRAM: 12.4 GBHeadroom: 3.6 GBTTFT: noticeable- • Tight VRAM fit — only 3.6 GB headroom left for context growth
ollama run qwen3:14b74tok/sEstimated
Quant: Q4_K_MContext: 2,048VRAM: 12.4 GBHeadroom: 3.6 GBTTFT: noticeable
- • Tight VRAM fit — only 3.6 GB headroom left for context growth
ollama run qwen3:14b74
tok/s
Estimated
Weights
8.40 GB
KV cache
1.75 GB
Activations
0.42 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~640 ms (noticeable)
Quant: Q4_K_MContext: 2,048VRAM: 12.9 GBHeadroom: 3.1 GBTTFT: noticeable- • Tight VRAM fit — only 3.1 GB headroom left for context growth
ollama run qwen2.5:14b74tok/sEstimated
Quant: Q4_K_MContext: 2,048VRAM: 12.9 GBHeadroom: 3.1 GBTTFT: noticeable
- • Tight VRAM fit — only 3.1 GB headroom left for context growth
ollama run qwen2.5:14b74
tok/s
Estimated
Weights
8.90 GB
KV cache
1.75 GB
Activations
0.45 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~640 ms (noticeable)
Quant: Q4_K_MContext: 2,048VRAM: 12.8 GBHeadroom: 3.2 GBTTFT: fast- • Tight VRAM fit — only 3.2 GB headroom left for context growth
431tok/sEstimated
Quant: Q4_K_MContext: 2,048VRAM: 12.8 GBHeadroom: 3.2 GBTTFT: fast
- • Tight VRAM fit — only 3.2 GB headroom left for context growth
431
tok/s
Estimated
Weights
8.60 GB
KV cache
1.96 GB
Activations
0.43 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~110 ms (fast)
Quant: Q4_K_MContext: 8,192VRAM: 29.8 GBHeadroom: 5.4 GBTTFT: noticeable- • Partial CPU offload: ~46% of layers run on CPU
40tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 29.8 GBHeadroom: 5.4 GBTTFT: noticeable
- • Partial CPU offload: ~46% of layers run on CPU
40
tok/s
Estimated
Weights
14.30 GB
KV cache
13.00 GB
Activations
0.72 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~1189 ms (noticeable)
Quant: Q4_K_MContext: 8,192VRAM: 32.1 GBHeadroom: 3.1 GBTTFT: noticeable- • Partial CPU offload: ~50% of layers run on CPU
ollama run gemma3:27b38tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 32.1 GBHeadroom: 3.1 GBTTFT: noticeable
- • Partial CPU offload: ~50% of layers run on CPU
ollama run gemma3:27b38
tok/s
Estimated
Weights
16.00 GB
KV cache
13.50 GB
Activations
0.81 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~1234 ms (noticeable)
Quant: Q4_K_MContext: 2,048VRAM: 24.5 GBHeadroom: 10.7 GBTTFT: noticeable- • Partial CPU offload: ~35% of layers run on CPU
ollama run qwen3:30b34tok/sEstimated
Quant: Q4_K_MContext: 2,048VRAM: 24.5 GBHeadroom: 10.7 GBTTFT: noticeable
- • Partial CPU offload: ~35% of layers run on CPU
ollama run qwen3:30b34
tok/s
Estimated
Weights
18.00 GB
KV cache
3.75 GB
Activations
0.90 GB
Runtime
1.80 GB
Time to first token (prefill, 512-token prompt): ~1371 ms (noticeable)
What if you upgraded?
Hypothetical scenarios. We re-ran the compatibility engine for each.
+32 GB system RAM
~$80–150
Doubles your CPU-offload working set. Helps when models don't quite fit in VRAM.
Unlocks: 41 new comfortable, 112 new tradeoff
- • Qwen 3 0.6B
- • Qwen 3 1.7B
- • Gemma 3 270M
- • SmolLM2 135M Instruct
Upgrade to NVIDIA RTX 2080 Ti 22GB (China-mod)
~$350
22 GB VRAM (vs your 16 GB) plus a bandwidth jump from ~960 GB/s to ~616 GB/s.
Unlocks: 68 new comfortable
- • Qwen 3 0.6B
- • Gemma 4 12B
- • Qwen3.5 9B
- • Qwen 3 1.7B
Add a second NVIDIA GeForce RTX 5080
~$1199
Tensor parallelism splits the model across both cards, effectively doubling VRAM. Bandwidth doesn't double — runs ~1.5× the single-card speed in practice.
Unlocks: 115 new comfortable
- • Qwen 3 0.6B
- • Qwen 3 30B-A3B
- • Qwen 2.5 Coder 32B Instruct
- • Qwen3.6 27B
Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.
Won't runtop 5 popular models
Need more memory than you have. Shown for orientation.
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
How to read these numbers
Measured here
Measured here - RunLocalAI ran this exact combo on owner hardware with public evidence.Source-backed
Source-backed / community - a reproduced public source supports the speed, but it is not labeled as owner-measured.Extrapolated
Extrapolated - predicted from a measured benchmark on similar-bandwidth hardware.Estimated
Estimated - formula based on VRAM bandwidth and model architecture; not a benchmark row.Want a specific benchmark we don't have? Email Contact support and we'll prioritize it.