FitMyLLM
NVIDIA· MAXWELL

NVIDIA Quadro K2200

Running LLMs on the Quadro K2200 — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
4 GB
BUDGET
BANDWIDTH
80.2
GB/S
MODELS Q4
97/449
22%
7B Q4 SPEED
~9
USABLE
▸ MODEL COVERAGE @ Q422% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~21
TOK/S
7B
3.9GB NEEDED
14B
7.9GB NEEDED
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
4 GB
BANDWIDTH
80.2 GB/s
FP16 COMPUTE
1.4 TFLOPS
TDP
68W
MEMORY
GDDR5
ARCHITECTURE
Maxwell
CUDA CORES
640
PCIE
Gen 2 x16
57
FAST MODELS · >30 TOK/S
Real-time chat speed
97
USABLE · >10 TOK/S
Comfortable for all tasks
97
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test Quadro K2200 (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· 97
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
713
TOK/S · 14% VRAM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
475
TOK/S · 14% VRAM
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
458
TOK/S · 14% VRAM
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
458
TOK/S · 14% VRAM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
251
TOK/S · 16% VRAM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
238
TOK/S · 16% VRAM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
210
TOK/S · 17% VRAM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
192
TOK/S · 17% VRAM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
192
TOK/S · 17% VRAM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
192
TOK/S · 17% VRAM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
183
TOK/S · 18% VRAM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
178
TOK/S · 18% VRAM
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
169
TOK/S · 18% VRAM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
128
TOK/S · 20% VRAM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
128
TOK/S · 20% VRAM
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
123
TOK/S · 20% VRAM
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
113
TOK/S · 21% VRAM
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
113
TOK/S · 21% VRAM
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
107
TOK/S · 21% VRAM
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
107
TOK/S · 21% VRAM
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
107
TOK/S · 21% VRAM
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
103
TOK/S · 22% VRAM
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
87
TOK/S · 24% VRAM
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
86
TOK/S · 24% VRAM
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
79
TOK/S · 25% VRAM
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
74
TOK/S · 26% VRAM
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
68
TOK/S · 27% VRAM
A
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
59
TOK/S · 29% VRAM
A
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
58
TOK/S · 29% VRAM
A
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
53
TOK/S · 31% VRAM
A
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
53
TOK/S · 31% VRAM
A
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
53
TOK/S · 31% VRAM
A
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
53
TOK/S · 31% VRAM
A
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
49
TOK/S · 32% VRAM
A
OPT 1.3B1.3B
OPT·2K CTX· CHAT
49
TOK/S · 32% VRAM
A
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
49
TOK/S · 32% VRAM
A
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
48
TOK/S · 33% VRAM
A
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
45
TOK/S · 34% VRAM
A
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
45
TOK/S · 34% VRAM
A
Qwen2.5-Coder-1.5B1.5B
QWEN·32K CTX· CHAT· TOOL_USE· CODING
43
TOK/S · 35% VRAM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your Quadro K2200.

WHAT THIS CARD IS WORTH

Quadro K2200 holds 97 of the models in our catalogue and is, in practice, a Q4_K_M card — the largest it takes is InternLM2 5B at Q4_K_M.

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
InternLM2 5B4.5B · Q4_K_M3.6 GB22 tok/sEST
Ministral 3 3B Reasoning4.25B · Q4_K_M3.3 GB23 tok/sEST
Qwen3 4B4B · Q4_K_M3.3 GB25 tok/sEST
Qwen3-4B Instruct 25074B · Q4_K_M3.3 GB25 tok/sEST
Qwen3-Embedding 4B4B · Q4_K_M3.3 GB25 tok/sEST
Nemotron 3 Nano 4B3.97B · Q4_K_M3.4 GB25 tok/sEST
Ministral 3 3B3.85B · Q5_K_M3.5 GB22 tok/sEST
phi-3-mini-4k 3.8B3.8B · Q4_K_M3.4 GB26 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
RTX 3070 Ti8 GB$49912.6 tok/s per $100
RTX 3060 Ti GDDR6X8 GB$39915.8 tok/s per $100
RTX 3070 Ti 8 GB GA1028 GB$59910.5 tok/s per $100
Arc A7508 GB$19910.1 tok/s per $100
Quadro K22004 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 K2200 — 4 GB VRAM.

QUADRO K2200 SPEC
BRAND
NVIDIA
VRAM
4 GB GDDR5
BANDWIDTH
80.2 GB/s
FP16 COMPUTE
1.4 TFLOPS
FP32 COMPUTE
1.4 TFLOPS
CUDA CORES
640
TDP
68 W
ARCHITECTURE
Maxwell
▸ AI CAPABILITY
97/ 449 models @ Q4

With 4 GB VRAM and 80.2 GB/s bandwidth, this GPU handles models up to 4.25B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~9 tok/s.

§ 01TOP MODELS FOR QUADRO K2200
97 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
Ministral 3 3B Reasoning4.25B3.1 GB15
Qwen 1.5 4B4B2.9 GB1612.6
Qwen3 4B4B2.9 GB1640.7
Qwen3-4B Instruct 25074B2.9 GB1637.2
Qwen3-Embedding 4B4B2.9 GB16
Nemotron 3 Nano 4B3.97B2.9 GB1632.0
Ministral 3 3B3.85B2.8 GB1721.4
Phi-3.5 Mini 3.8B3.82B2.8 GB1746.6
phi-3-mini-4k 3.8B3.8B2.8 GB1730.5
Phi-4-mini 3.8B3.8B2.8 GB1749.0
Cogito 3B3.61B2.7 GB1822.1
Falcon3-3B3.23B2.5 GB2025.7
granite-4.0-h-micro 3.2B3.2B2.4 GB2018.4
Llama-3.2-3B3.2B2.4 GB2017.9
Falcon-H1 3B3.15B2.4 GB2049.5
Qwen 2.5 3B3.1B2.4 GB2137.2
SmolLM3-3B3.1B2.4 GB2130.5
Ministral 3B3B2.3 GB2129.6
StarCoder2 3B3B2.3 GB219.5
Granite 4.1 3B3B2.3 GB2116.6