LocalLLMChecker

Workstation GPU · NVIDIA

What AI models can an RTX 6000 Ada Generation run?

48 GB puts the 70B class within comfortable reach and leaves room for the context window that makes those models worth running. At 960 GB/s there is enough bandwidth to keep generation responsive for anything that fits.

Memory48 GB
Bandwidth960 GB/s
Usable for a model47.2 GB
Runtime backendCUDA
ArchitectureAda Lovelace
Power300 W

The short answer

Assuming an 8k context window and default settings, these are the models worth downloading first.

Best all-rounder

Gemma 4 31B Instruct

Q8_0 · 34.67 GB · about 20.6 tokens/s

Best for code

Qwen3 32B

Q8_0 · 34.09 GB · about 21 tokens/s

Largest that still runs well

Qwen3 Next 80B A3B Instruct

81.32B parameters · Q4_K_M · about 200.2 tokens/s

Fastest useful answer

LFM2.5 8B A1B

about 248.7 tokens/s · 8.81 GB

See how fast it feels

RTX 6000 Ada Generation running Gemma 4 31B Instruct at Q8_0

Wait for the first word230 ms
Then writes at20.6 tok/s
Whole answer8.6 s

YouWhy does my model use more memory when the conversation gets longer?

Model

Simulated from our estimate at a 512-token question, not a recording. Assumes nothing else is competing for the GPU.

Where it sits

With 48 GB it belongs to the 48 GB tier. The 48 GB guide explains what that tier runs and where it runs out, and the bandwidth guide explains why 960 GB/s decides how fast it answers.

Our indicative price for this card is $6,800 new, reviewed 2026-09-01. That is 0.7 GB of memory and 14 GB/s of bandwidth for every $100. Prices move, so treat these as a place to start a search and not a quote.

Devices in a similar price band, fastest memory first:

DevicePriceMemoryBandwidthBest all-round pick
RTX PRO 6000 Blackwell $8,500 96 GB 1792 GB/s GPT OSS 120B, 242.6 t/s

Every model, scored on this device

Each row uses the highest-quality quantization that both fits and stays conversational. Speed is a single-stream estimate at 8k context.

Model Params Verdict Download Memory used Speed Max context
Gemma 4 31B Instruct
Gemma 		31.27B Runs well Q8_0 34.67 GB 20.6 t/s 32k
Qwen3 32B
Qwen 		32.76B Runs well Q8_0 34.09 GB 21 t/s 32k
Qwen3.5 27B
Qwen 		27.78B Runs great Q8_0 27.32 GB 26.3 t/s 256k
Granite 4.1 30B
Granite 		28.87B Runs well Q8_0 30.97 GB 23.1 t/s 64k
Qwen2.5 32B Instruct
Qwen 		32.76B Runs well Q8_0 34.09 GB 21 t/s 32k
Qwen2.5 Coder 32B Instruct
Qwen 		32.76B Runs well Q8_0 34.09 GB 21 t/s 32k
Llama 3.3 70B Instruct
Llama 		70.55B Runs well Q4_K_M 42.12 GB 17 t/s 16k
Qwen2.5 VL 72B Instruct
Qwen 		73.41B Runs well Q4_K_M 43.62 GB 16.4 t/s 16k
Qwen2.5 72B Instruct
Qwen 		72.71B Runs well Q4_K_M 43.22 GB 16.5 t/s 16k
Gemma 3 27B Instruct
Gemma 		27.43B Runs great Q8_0 28.68 GB 25 t/s 128k
Qwen3.5 35B A3B
Qwen 		35.95B (2.9B active) Runs great Q8_0 35.63 GB 134.5 t/s 256k
Gemma 4 26B A4B Instruct
Gemma 		25.81B (3.74B active) Runs great Q8_0 26.72 GB 90.6 t/s 256k
Qwen3 Coder Next
Qwen 		79.67B (3.19B active) Runs great Q4_K_M 45.22 GB 200.2 t/s 64k
GLM 4.7 Flash
GLM 		31.22B (3.66B active) Runs great Q8_0 31.33 GB 102.5 t/s 128k
Mixtral 8x7B Instruct V0.1
Mistral 		46.7B (12.88B active) Runs great Q6_K 36.97 GB 37.9 t/s 32k
Qwen3 Next 80B A3B Instruct
Qwen 		81.32B (3.19B active) Runs great Q4_K_M 46.15 GB 200.2 t/s 32k
Qwen3 30B A3B
Qwen 		30.53B (3.34B active) Runs great Q8_0 30.99 GB 101.9 t/s 32k
Mistral Small 24B Instruct 2501
Mistral 		23.57B Runs great Q8_0 24.35 GB 29.5 t/s 32k
Phi 3.5 MoE Instruct
Phi 		41.87B (6.64B active) Runs great Q8_0 42.71 GB 54.3 t/s 32k
Gemma 4 12B Instruct
Gemma 		11.96B Runs great Q8_0 13.55 GB 53.7 t/s 256k
Qwen3 14B
Qwen 		14.77B Runs great Q8_0 15.54 GB 46.8 t/s 32k
Llama 3 3 Nemotron Super 49B V1
Llama 		49.87B Runs well IQ4_XS 44.7 GB 16 t/s 8k
GPT OSS 20B
GPT-OSS 		20.91B (4.18B active) Runs great MXFP4 11.55 GB 171.5 t/s 128k
Qwen3 Coder 30B A3B Instruct
Qwen 		30.53B (3.34B active) Runs great Q8_0 30.99 GB 101.9 t/s 32k
Qwen2.5 14B Instruct
Qwen 		14.77B Runs great Q8_0 15.79 GB 46 t/s 32k
Qwen2.5 Coder 14B Instruct
Qwen 		14.77B Runs great Q8_0 15.79 GB 46 t/s 32k
Codestral 22B V0.1
Mistral 		22.25B Runs great Q8_0 24.04 GB 30 t/s 32k
Qwen3.5 9B
Qwen 		9.65B Runs great Q8_0 9.25 GB 79.9 t/s 256k
Phi 4
Phi 		14.66B Runs great Q8_0 16 GB 45.4 t/s 16k
Gemma 3 12B Instruct
Gemma 		12.19B Runs great Q8_0 13.27 GB 54.9 t/s 128k
Show the other 56 models
Gemma 4 E4B Instruct 8B Runs great Q8_0 8.42 GB 87.6 t/s 128k
Granite 4.1 8B 8.79B Runs great Q8_0 10.35 GB 71 t/s 128k
Qwen3 8B 8.19B Runs great Q8_0 9.02 GB 82 t/s 32k
Mistral Nemo Instruct 2407 12.25B Runs great Q8_0 13.15 GB 55.6 t/s 128k
Qwen2.5 VL 7B Instruct 8.29B Runs great Q8_0 8.49 GB 87.2 t/s 64k
Qwen2.5 7B Instruct 7.62B Runs great Q8_0 7.83 GB 95 t/s 32k
Llama 3.1 8B Instruct 8.03B Runs great Q8_0 8.83 GB 83.8 t/s 128k
DeepSeek Coder V2 Lite Instruct 15.71B Runs great Q8_0 15.91 GB 140.7 t/s 128k
Gemma 2 9B Instruct 9.24B Runs great Q8_0 11.5 GB 63.6 t/s 8k
Qwen2.5 Coder 7B Instruct 7.62B Runs great Q8_0 7.83 GB 95 t/s 32k
OLMo 3 7B Instruct 7.3B Runs great Q8_0 9.72 GB 75.8 t/s 64k
Internlm3 8B Instruct 8.8B Runs great Q8_0 8.97 GB 82.5 t/s 32k
Mistral 7B Instruct V0.3 7.25B Runs great Q8_0 8.45 GB 87.8 t/s 32k
Gemma 4 E2B Instruct 5.12B Runs great Q8_0 5.44 GB 138 t/s 128k
Qwen3.5 4B 4.66B Runs great Q8_0 5.27 GB 143.9 t/s 256k
Mistral 7B Instruct V0.2 7.24B Runs great Q8_0 8.44 GB 87.9 t/s 32k
DeepSeek Coder 7B Instruct V1.5 6.91B Runs great Q8_0 10.58 GB 69.4 t/s 4k
DeepSeek Coder 6.7B Instruct 6.74B Runs great Q8_0 10.94 GB 67 t/s 16k
LFM2.5 8B A1B 8.47B Runs great Q8_0 8.81 GB 248.7 t/s 64k
Qwen3 4B 4.02B Runs great Q8_0 5.46 GB 138.4 t/s 32k
Gemma 3 4B Instruct 4.3B Runs great Q8_0 4.91 GB 155.2 t/s 128k
Phi 4 Mini Instruct 3.84B Runs great Q8_0 5.17 GB 147.2 t/s 128k
Granite 4.1 3B 3.4B Runs great Q8_0 4.34 GB 177 t/s 128k
Qwen2.5 3B Instruct 3.09B Runs great Q8_0 3.68 GB 211.5 t/s 32k
Phi 3 Mini 4k Instruct 3.82B Runs great Q8_0 4.8 GB 159.4 t/s 4k
Llama 3.2 3B Instruct 3.21B Runs great Q8_0 4.42 GB 174.3 t/s 128k
LFM2.5 2.6B 2.7B Runs great Q8_0 3.14 GB 252.6 t/s 128k
Qwen3.5 2B 2.27B Runs great Q8_0 2.7 GB 299.5 t/s 256k
Gemma 2 2B Instruct 2.61B Runs great Q8_0 3.54 GB 221.3 t/s 8k
Starcoder2 3B 3.03B Runs great Q8_0 3.34 GB 238.1 t/s 16k
Qwen3 1.7B 2.03B Runs great Q8_0 3.22 GB 244.9 t/s 32k
DeepSeek R1 Distill Qwen 1.5B 1.78B Runs great Q8_0 2.07 GB 403.8 t/s 128k
Qwen2.5 1.5B Instruct 1.54B Runs great Q8_0 2.06 GB 405.3 t/s 32k
SmolLM2 1.7B 1.71B Runs great Q8_0 3.53 GB 221.3 t/s 8k
OLMo 2 0425 1B 1.48B Runs great Q8_0 2.6 GB 312.4 t/s 4k
Llama 3.2 1B Instruct 1.24B Runs great Q8_0 1.82 GB 478.2 t/s 128k
LFM2.5 1.2B Instruct 1.17B Runs great Q8_0 1.46 GB 631.4 t/s 64k
Gemma 3 1B Instruct 1B Runs great Q8_0 1.35 GB 679.9 t/s 32k
Qwen3.5 0.8B 0.87B Runs great Q8_0 1.28 GB 726.1 t/s 256k
Qwen3 0.6B 0.75B Runs great Q8_0 1.92 GB 436.8 t/s 32k
Qwen2.5 0.5B Instruct 0.49B Runs great Q8_0 0.88 GB 1220.7 t/s 32k
LFM2.5 350M 0.35B Runs great Q8_0 0.68 GB 1890.2 t/s 64k
GPT OSS 120B 116.83B Partial offload n/a 59.4 GB 53.1 t/s n/a
Qwen3.5 122B A10B 125.09B Partial offload n/a 70.61 GB 29.7 t/s n/a
GLM 4.5 Air 110.47B Partial offload n/a 63.61 GB 18.9 t/s n/a
MiniMax M2.5 228.7B Partial offload n/a 130.56 GB 10 t/s n/a
MiniMax M2.7 228.69B Partial offload n/a 130.55 GB 10 t/s n/a
Qwen3 235B A22B 235.09B Partial offload n/a 133.71 GB 5 t/s n/a
MiniMax M3 427.04B Partial offload n/a 240.84 GB 4.5 t/s n/a
GLM 4.5 358.34B Partial offload n/a 204.37 GB 3.4 t/s n/a
DeepSeek V3.2 685.4B Partial offload n/a 385.91 GB 2.6 t/s n/a
DeepSeek R1 684.53B Partial offload n/a 385.42 GB 2.6 t/s n/a
Kimi K2.6 1026.88B Partial offload n/a 577.78 GB 2.5 t/s n/a
Kimi K2 Instruct 1026.41B Partial offload n/a 577.52 GB 2.5 t/s n/a
GLM 5.1 753.86B Partial offload n/a 424.51 GB 2.4 t/s n/a
GLM 5.2 753.33B Partial offload n/a 424.21 GB 2.3 t/s n/a

What it will not run

0 of the 86 architectures we track are out of reach here, even at 2-bit precision. Another 14 run by splitting the model between the card and system RAM, which works but drops generation to single digits for most dense models.

If you want more headroom

The next steps up in memory, in order. More memory changes which models load at all; more bandwidth changes how fast they answer, so the two are worth weighing separately.

DeviceMemoryBandwidthModels that fit
Ryzen AI Max+ 395 64GB 64 GB 256 GB/s 66 Check price
Jetson AGX Orin 64GB 64 GB 204.8 GB/s 66 Check price
RTX PRO 6000 Blackwell 96 GB 1792 GB/s 76 Check price
Ryzen AI Max+ 395 96GB 96 GB 256 GB/s 69 Check price

Price links go to an Amazon search for the model name and are affiliate links: if you buy through one, we earn a commission at no cost to you. We do not take payment for placement, and the ordering above is by memory capacity alone.

Other devices with 48 GB

Same capacity, different speed. Once a model fits, bandwidth is what separates these.

Radeon PRO W7900

864 GB/s · AMD

RTX A6000

768 GB/s · NVIDIA

Apple M4 Max 48GB

546 GB/s · Apple

Apple M3 Max 48GB

400 GB/s · Apple

Apple M4 Pro 48GB

273 GB/s · Apple

Questions people ask about an RTX 6000 Ada Generation

How much VRAM does an RTX 6000 Ada Generation have?

The RTX 6000 Ada Generation has 48 GB of VRAM, of which roughly 47.2 GB is available to a model once the operating system takes its share.

Can an RTX 6000 Ada Generation run a 70B model?

Yes. At Q4_K_M a 70B model fits in 42.12 GB and generates around 17 tokens per second.

What is the best local model for an RTX 6000 Ada Generation?

Gemma 4 31B Instruct at Q8_0 is the strongest all-round fit: 34.67 GB used, around 20.6 tokens per second at 8k context.