RTX 3080

NVIDIA GeForce RTX 30 (Ampere)

VRAM: 10 GB
Memory bandwidth: 760 GB/s
fp16 compute: 119.1 TFLOPS
Class: Consumer

Figures assume this GPU plus 32 GB of system RAM (a typical desktop pairing). "What runs on it" is judged at a 8,192-token context. Speeds are estimates, not measurements.

What runs on it

Model	Sweet-spot quant	Est. speed	Community
DeepSeek-R1-Distill-Llama-8B deepseek-ai	Q4_K_M Runs fully on GPU @ 8K ctx	76–101 tok/s (estimate)	no community data
DeepSeek-R1-Distill-Qwen-1.5B deepseek-ai	Q8_0 Runs fully on GPU @ 8K ctx	214–286 tok/s (estimate)	no community data
DeepSeek-R1-Distill-Qwen-7B deepseek-ai	Q8_0 Runs fully on GPU @ 8K ctx	53–71 tok/s (estimate)	no community data
Kimi-VL-A3B-Instruct moonshotai	IQ4_XS Runs fully on GPU @ 8K ctx		no community data
Llama-3.1-8B-Instruct meta-llama	Q4_K_M Runs fully on GPU @ 8K ctx	76–101 tok/s (estimate)	no community data
Phi-4-mini-instruct microsoft	Q8_0 Runs fully on GPU @ 8K ctx	88–118 tok/s (estimate)	no community data
Qwen2.5-7B-Instruct Qwen	Q8_0 Runs fully on GPU @ 8K ctx	53–71 tok/s (estimate)	no community data
Qwen3-8B Qwen	Q4_K_M Runs fully on GPU @ 8K ctx	73–98 tok/s (estimate)	no community data
Qwen3-Embedding-0.6B Qwen	Q8_0 Runs fully on GPU @ 8K ctx	289–385 tok/s (estimate)	no community data
Qwen3-Embedding-4B Qwen	Q4_K_M Runs fully on GPU @ 8K ctx	123–164 tok/s (estimate)	no community data
Qwen3-Embedding-8B Qwen	Q4_K_M Runs fully on GPU @ 8K ctx	77–103 tok/s (estimate)	no community data
Qwen3-Reranker-0.6B Qwen	BF16 Runs fully on GPU @ 8K ctx	214–285 tok/s (estimate)	no community data
Qwen3-Reranker-4B Qwen	BF16 Runs fully on GPU @ 8K ctx	49–66 tok/s (estimate)	no community data
SmolLM3-3B HuggingFaceTB	BF16 Runs fully on GPU @ 8K ctx	67–90 tok/s (estimate)	no community data
dots.ocr rednote-hilab	BF16 Runs fully on GPU @ 8K ctx	72–96 tok/s (estimate)	no community data
gemma-4-12B-it google	Q4_K_M Runs fully on GPU @ 8K ctx	57–76 tok/s (estimate)	no community data
gemma-4-E2B-it google	Q8_0 Runs fully on GPU @ 8K ctx	89–119 tok/s (estimate)	no community data
gemma-4-E4B-it google	Q8_0 Runs fully on GPU @ 8K ctx	55–73 tok/s (estimate)	no community data
DeepSeek-R1-Distill-Qwen-14B deepseek-ai	Q8_0 CPU offload @ 8K ctx · ~22 GPU layers	4–6 tok/s (estimate)	no community data
DeepSeek-R1-Distill-Qwen-32B deepseek-ai	Q8_0 CPU offload @ 8K ctx · ~10 GPU layers	1–2 tok/s (estimate)	no community data
Devstral-Small-2-24B-Instruct-2512 mistralai	Q8_0 CPU offload @ 8K ctx · ~11 GPU layers	2–3 tok/s (estimate)	no community data
GLM-4.7-Flash zai-org	Q8_0 CPU offload @ 8K ctx · ~11 GPU layers		no community data
Kimi-Dev-72B moonshotai	IQ4_XS CPU offload @ 8K ctx · ~10 GPU layers	1 tok/s (estimate)	no community data
Mistral-Small-3.2-24B-Instruct-2506 mistralai	Q8_0 CPU offload @ 8K ctx · ~11 GPU layers	2–3 tok/s (estimate)	no community data
Phi-4-reasoning microsoft	Q8_0 CPU offload @ 8K ctx · ~18 GPU layers	4–5 tok/s (estimate)	no community data
Qwen2.5-Omni-7B Qwen	BF16 CPU offload @ 8K ctx · ~10 GPU layers	3–4 tok/s (estimate)	no community data
Qwen2.5-VL-32B-Instruct Qwen	Q8_0 CPU offload @ 8K ctx · ~10 GPU layers	1–2 tok/s (estimate)	no community data
Qwen3-14B Qwen	Q8_0 CPU offload @ 8K ctx · ~19 GPU layers	4–6 tok/s (estimate)	no community data
Qwen3-30B-A3B-Instruct-2507 Qwen	Q8_0 CPU offload @ 8K ctx · ~10 GPU layers		no community data
Qwen3-32B Qwen	Q8_0 CPU offload @ 8K ctx · ~10 GPU layers	1–2 tok/s (estimate)	no community data
Qwen3-Coder-30B-A3B-Instruct Qwen	Q8_0 CPU offload @ 8K ctx · ~10 GPU layers		no community data
Qwen3-Omni-30B-A3B-Instruct Qwen	Q4_K_M CPU offload @ 8K ctx · ~20 GPU layers		no community data
Qwen3-Reranker-8B Qwen	BF16 CPU offload @ 8K ctx · ~17 GPU layers	4–6 tok/s (estimate)	no community data
Qwen3.6-27B Qwen	Q8_0 CPU offload @ 8K ctx · ~13 GPU layers	2 tok/s (estimate)	no community data
Qwen3.6-35B-A3B Qwen	Q8_0 CPU offload @ 8K ctx · ~7 GPU layers		no community data
gemma-4-26B-A4B-it google	Q8_0 CPU offload @ 8K ctx · ~8 GPU layers		no community data
gemma-4-31B-it google	Q8_0 CPU offload @ 8K ctx · ~11 GPU layers	1–2 tok/s (estimate)	no community data
gpt-oss-20b openai	F16 CPU offload @ 8K ctx · ~15 GPU layers		no community data
phi-4 microsoft	Q8_0 CPU offload @ 8K ctx · ~18 GPU layers	4–5 tok/s (estimate)	no community data

"Est. speed" is a modelled range labelled estimate (D8) for generation (decode) throughput. "Community" shows the median of approved user-submitted reports on this GPU class only where enough exist — never an estimate. "pp" is measured prompt-processing (ingestion) throughput from approved community reports; rows without a measurement show none.

What can I run on a RTX 3080?

On this GPU, 12 catalog models run fully on the GPU at an 8,192-token context. The most capable is Kimi-VL-A3B-Instruct at IQ4_XS (needs ~9.7 GiB). Pick a smaller model or a lower quant for more headroom.

Biggest model: Kimi-VL-A3B-Instruct at IQ4_XS

llama-server -m Kimi-VL-A3B-Instruct.i1-IQ4_XS.gguf -c 8192 -ngl 999

Derived from the fit engine at an 8,192-token context. See more answer packs.

Add a second RTX 3080?

A second RTX 3080 pools VRAM: 2 × 10 GB = 20 GB combined. Bigger models can then load because their weights split across both cards — but a second card does not make generation proportionally faster (see the reality check below).

A second RTX 3080 adds 10 GB of VRAM for about $325 — ≈$32.50/GB of added VRAM (used price as of 2026-07-18 — BestValueGPU (eBay used price tracker)).

14 more catalog models could newly fit fully in the combined 20 GB at a 8,192-token context — for example:

GLM-4.7-Flash at Q4_K_M · zai-org
Qwen3.6-35B-A3B at IQ4_XS · Qwen
gemma-4-26B-A4B-it at Q4_K_M · google
Qwen3.6-27B at Q4_K_M · Qwen
Qwen3-Coder-30B-A3B-Instruct at IQ4_XS · Qwen

Estimate — assumes the model's weights split across both cards (a layer split, as llama.cpp does by default). This is a combined-VRAM projection, not a measured or verdict-chipped result: the fit engine treats two cards as one summed memory pool and does not model the link between them. Check your exact model and context in the calculator.

The honest reality of a second card

Bandwidth doesn't add. Two cards give more VRAM, not more memory bandwidth per token. Token generation is bandwidth-bound, so a layer-split model decodes at roughly one card's speed — not double.
Layer-split vs tensor-parallel. The common desktop setup (llama.cpp) splits layers across cards and runs them in sequence, so one GPU works at a time. True tensor-parallel serving (e.g. vLLM) can use both at once, but wants matched cards and a fast interconnect.
pp vs tg. Prompt processing (pp) can gain more from a second card than token generation (tg); raw decode throughput barely moves. Don't expect a 2× tok/s jump.
PCIe / NUMA. Cards talk over PCIe (or NVLink where supported), far slower than on-card VRAM. A layer split crosses it about once per token so the hit is small; tensor-parallel crosses it constantly, and cards on different CPU sockets (NUMA) add latency.
Power & PSU. A second card roughly doubles GPU power draw — check PSU headroom, connectors, slot spacing and airflow before buying.

Read: Multi-GPU for local LLMs — when a second card is worth it →

Derived from the fit engine at a 8,192-token context, comparing a single 10 GB card against a summed 20 GB two-card pool.

Price history

Prices are point-in-time observations, not live quotes.

new

Sep 2020 · $699 new — Forbes launch coverage

Not enough history yet — trends appear once at least three dated observations are recorded.

used

Jul 2026 · $325 used — BestValueGPU (eBay used price tracker)

Not enough history yet — trends appear once at least three dated observations are recorded.