RTX A6000

NVIDIA RTX Workstation (Ampere)

VRAM
48 GB
Memory bandwidth
768 GB/s
fp16 compute
154.8 TFLOPS
Class
Workstation

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 Q8_0 Runs fully on GPU @ 8K ctx 48–64 tok/s (estimate) no community data
DeepSeek-R1-Distill-Qwen-1.5B deepseek-ai Q8_0 Runs fully on GPU @ 8K ctx 216–289 tok/s (estimate) no community data
DeepSeek-R1-Distill-Qwen-14B deepseek-ai Q8_0 Runs fully on GPU @ 8K ctx 27–35 tok/s (estimate) no community data
DeepSeek-R1-Distill-Qwen-32B deepseek-ai Q8_0 Runs fully on GPU @ 8K ctx 12–17 tok/s (estimate) no community data
DeepSeek-R1-Distill-Qwen-7B deepseek-ai Q8_0 Runs fully on GPU @ 8K ctx 54–72 tok/s (estimate) no community data
Devstral-Small-2-24B-Instruct-2512 mistralai Q8_0 Runs fully on GPU @ 8K ctx 17–23 tok/s (estimate) no community data
GLM-4.7-Flash zai-org Q8_0 Runs fully on GPU @ 8K ctx no community data
Hunyuan-A13B-Instruct tencent IQ4_XS Runs fully on GPU @ 8K ctx no community data
Kimi-Dev-72B moonshotai IQ4_XS Runs fully on GPU @ 8K ctx 11–14 tok/s (estimate) no community data
Kimi-VL-A3B-Instruct moonshotai BF16 Runs fully on GPU @ 8K ctx no community data
Llama-3.1-8B-Instruct meta-llama Q8_0 Runs fully on GPU @ 8K ctx 48–64 tok/s (estimate) no community data
Llama-3.3-70B-Instruct meta-llama Q4_K_M Runs fully on GPU @ 8K ctx 10–14 tok/s (estimate) no community data
Mistral-Small-3.2-24B-Instruct-2506 mistralai Q8_0 Runs fully on GPU @ 8K ctx 17–23 tok/s (estimate) no community data
Phi-4-mini-instruct microsoft Q8_0 Runs fully on GPU @ 8K ctx 89–119 tok/s (estimate) no community data
Phi-4-reasoning microsoft Q8_0 Runs fully on GPU @ 8K ctx 27–36 tok/s (estimate) no community data
Qwen2.5-7B-Instruct Qwen Q8_0 Runs fully on GPU @ 8K ctx 54–72 tok/s (estimate) no community data
Qwen2.5-Omni-7B Qwen BF16 Runs fully on GPU @ 8K ctx 20–27 tok/s (estimate) no community data
Qwen2.5-VL-32B-Instruct Qwen Q8_0 Runs fully on GPU @ 8K ctx 12–17 tok/s (estimate) no community data
Qwen3-14B Qwen Q8_0 Runs fully on GPU @ 8K ctx 27–36 tok/s (estimate) no community data
Qwen3-30B-A3B-Instruct-2507 Qwen Q8_0 Runs fully on GPU @ 8K ctx no community data
Qwen3-32B Qwen Q8_0 Runs fully on GPU @ 8K ctx 12–17 tok/s (estimate) no community data
Qwen3-8B Qwen Q8_0 Runs fully on GPU @ 8K ctx 46–62 tok/s (estimate) no community data
Qwen3-Coder-30B-A3B-Instruct Qwen Q8_0 Runs fully on GPU @ 8K ctx no community data
Qwen3-Embedding-0.6B Qwen Q8_0 Runs fully on GPU @ 8K ctx 292–389 tok/s (estimate) no community data
Qwen3-Embedding-4B Qwen Q4_K_M Runs fully on GPU @ 8K ctx 124–166 tok/s (estimate) no community data
Qwen3-Embedding-8B Qwen Q4_K_M Runs fully on GPU @ 8K ctx 78–104 tok/s (estimate) no community data
Qwen3-Omni-30B-A3B-Instruct Qwen Q4_K_M Runs fully on GPU @ 8K ctx no community data
Qwen3-Reranker-0.6B Qwen BF16 Runs fully on GPU @ 8K ctx 216–288 tok/s (estimate) no community data
Qwen3-Reranker-4B Qwen BF16 Runs fully on GPU @ 8K ctx 50–66 tok/s (estimate) no community data
Qwen3-Reranker-8B Qwen BF16 Runs fully on GPU @ 8K ctx 26–35 tok/s (estimate) no community data
Qwen3.6-27B Qwen Q8_0 Runs fully on GPU @ 8K ctx 15–20 tok/s (estimate) no community data
Qwen3.6-35B-A3B Qwen Q8_0 Runs fully on GPU @ 8K ctx no community data
SmolLM3-3B HuggingFaceTB BF16 Runs fully on GPU @ 8K ctx 68–91 tok/s (estimate) no community data
dots.ocr rednote-hilab BF16 Runs fully on GPU @ 8K ctx 73–97 tok/s (estimate) no community data
gemma-4-12B-it google Q8_0 Runs fully on GPU @ 8K ctx 34–45 tok/s (estimate) no community data
gemma-4-26B-A4B-it google Q8_0 Runs fully on GPU @ 8K ctx no community data
gemma-4-31B-it google Q8_0 Runs fully on GPU @ 8K ctx 13–18 tok/s (estimate) no community data
gemma-4-E2B-it google Q8_0 Runs fully on GPU @ 8K ctx 90–120 tok/s (estimate) no community data
gemma-4-E4B-it google Q8_0 Runs fully on GPU @ 8K ctx 55–74 tok/s (estimate) no community data
gpt-oss-20b openai F16 Runs fully on GPU @ 8K ctx no community data
phi-4 microsoft Q8_0 Runs fully on GPU @ 8K ctx 27–36 tok/s (estimate) no community data
gpt-oss-120b openai F16 CPU offload @ 8K ctx · ~24 GPU layers 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 A6000?

On this GPU, 12 catalog models run fully on the GPU at an 8,192-token context. The most capable is Hunyuan-A13B-Instruct at IQ4_XS (needs ~46.2 GiB). Pick a smaller model or a lower quant for more headroom.

Biggest model: Hunyuan-A13B-Instruct at IQ4_XS 

llama-server -m tencent_Hunyuan-A13B-Instruct-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 A6000?

A second RTX A6000 pools VRAM: 2 × 48 GB = 96 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 A6000 adds 48 GB of VRAM for about $3,500 — ≈$72.92/GB of added VRAM (used price as of 2026-07-18 — GPUDojo).

2 more catalog models could newly fit fully in the combined 96 GB at a 8,192-token context — for example:

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 48 GB card against a summed 96 GB two-card pool.

Price history

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

new

  • Dec 2020 · $4,650 new — WCCFtech (channel MSRP) (medium confidence)

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

used

  • Jul 2026 · $3,500 used — GPUDojo

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

selected to compare · pick at least 2

Compare ()