FitMyLLM
NVIDIA· Pascal

NVIDIA Quadro P5200 Max-Q

Running LLMs on the Quadro P5200 Max-Q — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
16 GB
Mid-range
Bandwidth
230.9
GB/s
Models Q4
277/498
56%
7B Q4 speed
~26
Good
Model coverage at Q456% of all
Estimated speed· By model size at Q4

Average speeds at Q4 quantization. Actual performance varies by model architecture and context length.

3B
~62
tok/s
7B
~26
tok/s
14B
~13
tok/s
32B
—
18.0GB needed
70B
—
39.4GB needed
Measured rig reports

We rent the machine and time every model on it: decode, VRAM peak, concurrency, watts and cost per million tokens. Yours may already be one of them — and two are free to read in full.

See the reports →
Specifications
VRAM
16 GB
Bandwidth
230.9 GB/s
FP16 Compute
0.1 TFLOPS
TDP
100W
Memory
GDDR5
Architecture
Pascal
CUDA Cores
2,560
130
Fast models · >30 tok/s
Real-time chat speed
275
Usable · >10 tok/s
Comfortable for all tasks
277
Total compatible
Fit in VRAM at Q4
Don’t want to buy?

Test Quadro P5200 Max-Q (or anything bigger) without committing. Pay by the second, cancel anytime.

Spin up in ~60s. Pay by the second. Cancel anytime.

Some links are affiliate links — we may earn a small commission at no extra cost to you. This helps keep FitMyLLM free and independent.

Compatible models· 277
S
Falcon-H1R Tiny 90M0.09B
Falcon·256K ctx· chat· reasoning
2052
tok/s · 3% VRAM
›
S
SmolLM2 135M0.135B
Smollm·2K ctx· chat
1368
tok/s · 4% VRAM
›
S
nomic-embed-text-v1.5 100M0.14B
Embedding·8K ctx· chat
1319
tok/s · 4% VRAM
›
S
GPT-2 124M0.14B
Gpt2·1K ctx· chat
1319
tok/s · 4% VRAM
›
S
SmolVLM 256M0.256B
Smollm·8K ctx· chat· vision
722
tok/s · 4% VRAM
›
S
Gemma 3 270M0.27B
Gemma·32K ctx· chat
684
tok/s · 4% VRAM
›
S
Snowflake Arctic Embed M v2.00.305B
Embedding·8K ctx· embedding· multilingual
606
tok/s · 4% VRAM
›
S
bge-large-en-v1.5 335M0.335B
Embedding·1K ctx· chat
551
tok/s · 4% VRAM
›
S
mxbai-embed-large-v10.335B
Embedding·1K ctx· embedding
551
tok/s · 4% VRAM
›
S
Snowflake Arctic Embed L0.335B
Embedding·1K ctx· embedding
551
tok/s · 4% VRAM
›
S
LFM2 350M0.35B
Lfm·125K ctx· chat· tool_use
528
tok/s · 4% VRAM
›
S
SmolLM2 360M0.36B
Smollm·8K ctx· chat
513
tok/s · 4% VRAM
›
S
GPT-2 Medium 345M0.38B
Gpt2·1K ctx· chat
486
tok/s · 5% VRAM
›
S
Qwen 2.5 0.5B0.5B
Qwen·32K ctx· chat
369
tok/s · 5% VRAM
›
S
SmolVLM 500M0.5B
Smollm·8K ctx· chat· vision
369
tok/s · 5% VRAM
›
S
Falcon-H1 0.5B0.52B
Falcon·128K ctx· chat
355
tok/s · 5% VRAM
›
S
BGE-M30.568B
Embedding·8K ctx· embedding
325
tok/s · 5% VRAM
›
S
Snowflake Arctic Embed L v2.00.568B
Embedding·8K ctx· embedding· multilingual
325
tok/s · 5% VRAM
›
S
Qwen3-Embedding 0.6B0.6B
Embedding·32K ctx· embedding· multilingual
308
tok/s · 5% VRAM
›
S
Falcon-H1R Tiny 0.6B0.6B
Falcon·32K ctx· chat· reasoning
308
tok/s · 5% VRAM
›
S
Falcon Perception 0.6B0.6B
Falcon·4K ctx· vision
308
tok/s · 5% VRAM
›
S
Qwen 1.5 0.5B0.62B
Qwen·32K ctx· chat
298
tok/s · 5% VRAM
›
S
LFM2 700M0.74B
Lfm·125K ctx· chat· tool_use
250
tok/s · 6% VRAM
›
S
Qwen3 0.6B0.75B
Qwen·32K ctx· chat· reasoning
246
tok/s · 6% VRAM
›
S
GPT-2 Large 774M0.81B
Gpt2·1K ctx· chat
228
tok/s · 6% VRAM
›
S
Qwen 3.5 0.8B0.87B
Qwen·256K ctx· chat· coding· multilingual
212
tok/s · 6% VRAM
›
S
InternVL3 1B0.94B
Other·32K ctx· chat· vision
197
tok/s · 7% VRAM
›
S
MiniCPM5 1B1.08B
Minicpm·128K ctx· chat· reasoning· multilingual
171
tok/s · 7% VRAM
›
S
K2 Horizon 0.9B1.08B
Other·128K ctx· chat· coding· reasoning
171
tok/s · 7% VRAM
›
S
TinyLlama 1.1B1.1B
Llama·2K ctx· chat
168
tok/s · 7% VRAM
›
S
LFM2.5-1.2B-Thinking1.2B
Lfm·122K ctx· chat· reasoning· tool_use
154
tok/s · 8% VRAM
›
S
Llama-3.2-1B1.2B
Llama·4K ctx· chat
154
tok/s · 8% VRAM
›
S
LFM2 1.2B1.2B
Lfm·125K ctx· chat· tool_use· multilingual
154
tok/s · 8% VRAM
›
S
Zamba2 1.2B1.2B
Other·4K ctx· chat
154
tok/s · 8% VRAM
›
S
Ling 3.0 Tiny7.89BMoE
Other·128K ctx· chat· coding· reasoning
142
tok/s · 33% VRAM
›
S
EXAONE-4.0-1.2B1.3B
Exaone·64K ctx· chat
142
tok/s · 8% VRAM
›
S
OPT 1.3B1.3B
Opt·2K ctx· chat
142
tok/s · 8% VRAM
›
S
MiniCPM-V 4.61.3B
Other·256K ctx· chat· vision
142
tok/s · 8% VRAM
›
S
DeepSeek Coder 1.3B1.35B
Deepseek·16K ctx· coding
137
tok/s · 8% VRAM
›
S
Phi-1 1.3B1.42B
Phi·2K ctx· coding
130
tok/s · 8% VRAM
›
Next step

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Quadro P5200 Max-Q.

What this card is worth

Quadro P5200 Max-Q holds 277 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is ERNIE 4.5 21B A3B at IQ4_XS.

Tokens/sec per $100
—
8B at Q4_K_M, so cards compare like for like
VRAM per $100
—
what memory costs on this card
The biggest it takes
ERNIE 4.5 21B A3B21.95B · IQ4_XS13.9 GB80 tok/sest
GPT-OSS 20B21B · Q4_K_S13.8 GB64 tok/sest
Reka Flash 321B · Q4_K_S14.2 GB11 tok/sest
Reka Flash 3.121B · Q4_K_S14.2 GB11 tok/sest
InternLM2 20B19.8B · Q4_K_M14.3 GB11 tok/sest
InternLM2.5 20B19.8B · Q4_K_M14.3 GB11 tok/sest
Ling-lite 16.8B16.8B · Q5_K_M13.6 GB78 tok/sest
DeepSeek V2 Lite 16B16B · Q5_K_M12.9 GB78 tok/sest
Against cards of similar memory
M1 Max (32GB)21 GB$1,4993.0 tok/s per $100
M2 Max (32GB)21 GB$1,7992.5 tok/s per $100
M2 Pro (32GB)21 GB$1,4991.9 tok/s per $100
M4 (32GB)21 GB$1,1991.7 tok/s per $100
Quadro P5200 Max-Q16 GB——

Neighbours in memory rather than in price: memory decides whether a card can do the job at all, so two cards of the same size at different prices is the comparison you are making. Ordered by memory, then bandwidth — not by the value column, which is worked out from bandwidth and price alone and therefore rewards a cheap card whatever its software stack does to that bandwidth in practice. Read it as one input, not as a ranking.

Device under test

NVIDIA Quadro P5200 Max-Q — 16 GB VRAM.

Quadro P5200 Max-Q spec
Brand
NVIDIA
VRAM
16 GB GDDR5
Bandwidth
230.9 GB/s
FP16 Compute
0.1 TFLOPS
FP32 Compute
8 TFLOPS
CUDA Cores
2,560
TDP
100 W
Architecture
Pascal
AI capability
277/ 498 models @ Q4

With 16 GB VRAM and 230.9 GB/s bandwidth, this GPU handles models up to 19.8B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~26 tok/s.

Rent it — live prices
Checking what the clouds are charging…
Top models for Quadro P5200 Max-Q
277 fit · showing 20
ModelSizeVRAM Q4tok/sAvg
InternLM2 20B19.8B12.6 GB945.1
InternLM2.5 20B19.8B12.6 GB950.9
Ling-lite 16.8B16.8B10.8 GB77—
DeepSeek V2 Lite 16B16B10.3 GB7738.0
StarCoder2 15B15.96B10.2 GB1226.5
DeepSeek-Coder-V2-Lite 15.7B15.7B10.1 GB7743.0
DeepSeek-VL2 Small 16B15.7B10.1 GB7743.1
StarCoder 15B15.5B10.0 GB1221.0
InternVL3 14B15.12B9.7 GB1238.1
Phi-4-reasoning-vision 15B15B9.7 GB1242.8
DeepSeek R1 Distill Qwen 14B14.8B9.5 GB1243.9
DeepCoder 14B14.8B9.5 GB1236.0
Qwen2.5-Coder-14B14.8B9.5 GB1241.3
Qwen2.5-14B14.8B9.5 GB1241.3
Qwen3 14B14.8B9.5 GB1245.7
phi-4 14B14.66B9.4 GB1333.7
Phi-4-reasoning 14B14.66B9.4 GB1333.7
Phi-4-reasoning-plus 14B14.66B9.4 GB1372.2
Ministral 3 14B14B9.0 GB1323.9
Phi-3-medium-14b14B9.0 GB1333.7