FitMyLLM
NVIDIA· TURING

NVIDIA RTX 2080 SUPER

Running LLMs on the RTX 2080 SUPER — the long read: which models fit at which quantisation, and the settings worth changing. · Or what else $280 buys

VRAM
8 GB
ENTRY-LEVEL
BANDWIDTH
496
GB/S
MODELS Q4
201/449
45%
7B Q4 SPEED
~57
BLAZING
▸ MODEL COVERAGE @ Q445% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~132
TOK/S
7B
~57
TOK/S
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
8 GB
BANDWIDTH
496 GB/s
FP16 COMPUTE
22.4 TFLOPS
TDP
250W
MEMORY
GDDR6
ARCHITECTURE
Turing
CUDA CORES
3,072
TENSOR CORES
384
PCIE
Gen 3 x16
MSRP
$280
201
FAST MODELS · >30 TOK/S
Real-time chat speed
201
USABLE · >10 TOK/S
Comfortable for all tasks
201
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test RTX 2080 SUPER (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· 201
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
4409
TOK/S · 7% VRAM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
2939
TOK/S · 7% VRAM
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
2834
TOK/S · 7% VRAM
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
2834
TOK/S · 7% VRAM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
1550
TOK/S · 8% VRAM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
1470
TOK/S · 8% VRAM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
1301
TOK/S · 8% VRAM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
1184
TOK/S · 9% VRAM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
1184
TOK/S · 9% VRAM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
1184
TOK/S · 9% VRAM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
1134
TOK/S · 9% VRAM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
1102
TOK/S · 9% VRAM
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
1044
TOK/S · 9% VRAM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
794
TOK/S · 10% VRAM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
794
TOK/S · 10% VRAM
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
763
TOK/S · 10% VRAM
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
699
TOK/S · 10% VRAM
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
699
TOK/S · 10% VRAM
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
661
TOK/S · 11% VRAM
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
661
TOK/S · 11% VRAM
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
661
TOK/S · 11% VRAM
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
640
TOK/S · 11% VRAM
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
536
TOK/S · 12% VRAM
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
529
TOK/S · 12% VRAM
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
490
TOK/S · 12% VRAM
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
456
TOK/S · 13% VRAM
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
422
TOK/S · 13% VRAM
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
367
TOK/S · 14% VRAM
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
361
TOK/S · 15% VRAM
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
331
TOK/S · 15% VRAM
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
331
TOK/S · 15% VRAM
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
331
TOK/S · 15% VRAM
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
331
TOK/S · 15% VRAM
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
305
TOK/S · 16% VRAM
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
305
TOK/S · 16% VRAM
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
305
TOK/S · 16% VRAM
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
294
TOK/S · 16% VRAM
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
279
TOK/S · 17% VRAM
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
279
TOK/S · 17% VRAM
S
LFM2 8B A1B8.3BMoE
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
265
TOK/S · 70% VRAM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your RTX 2080 SUPER.

WHAT THIS CARD IS WORTH

RTX 2080 SUPER holds 201 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is Falcon3-10B at IQ4_XS.

TOKENS/SEC PER $100
19.3
8B at Q4_K_M, so cards compare like for like
VRAM PER $100
2.86 GB
what memory costs on this card
THE BIGGEST IT TAKES
Falcon3-10B10.3B · IQ4_XS7.1 GB42 tok/sEST
Bamba 9B v29.78B · Q4_K_M7.2 GB41 tok/sEST
RecurrentGemma 9B9.63B · Q4_K_M6.8 GB41 tok/sEST
glm-4-9b9.4B · Q4_K_M6.6 GB42 tok/sEST
Yi 1.5 9B9B · Q4_K_M6.5 GB44 tok/sEST
Yi Coder 9B9B · Q4_K_M6.5 GB44 tok/sEST
Ministral 3 8B8.92B · Q4_K_M6.6 GB44 tok/sEST
Ministral 3 8B Reasoning8.92B · Q4_K_M6.6 GB44 tok/sEST
AGAINST CARDS OF SIMILAR MEMORY
M3 Pro (18GB)12 GB$1,5991.5 tok/s per $100
M1 Pro (16GB)11 GB$9992.9 tok/s per $100
M2 Pro (16GB)11 GB$1,2992.2 tok/s per $100
M4 (16GB)11 GB$4994.0 tok/s per $100
RTX 2080 SUPER8 GB$28019.3 tok/s per $100

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 RTX 2080 SUPER — 8 GB VRAM.

RTX 2080 SUPER SPEC
BRAND
NVIDIA
VRAM
8 GB GDDR6
BANDWIDTH
496 GB/s
FP16 COMPUTE
22.4 TFLOPS
FP32 COMPUTE
11.2 TFLOPS
CUDA CORES
3,072
TENSOR CORES
384
TDP
250 W
ARCHITECTURE
Turing
MSRP
$280
▸ AI CAPABILITY
201/ 449 models @ Q4

With 8 GB VRAM and 496 GB/s bandwidth, this GPU handles models up to 9.63B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~57 tok/s.

RENT IT — LIVE PRICES
Checking what the clouds are charging…
§ 01TOP MODELS FOR RTX 2080 SUPER
201 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
RecurrentGemma 9B9.63B6.4 GB4135.0
glm-4-9b9.4B6.2 GB4220.5
gemma-2-9b9.2B6.1 GB4330.2
Yi 1.5 9B9B6.0 GB4430.3
Yi Coder 9B9B6.0 GB4435.8
Ministral 3 8B8.92B5.9 GB4425.7
Ministral 3 8B Reasoning8.92B5.9 GB44
NVIDIA-Nemotron-Nano-9B-v28.9B5.9 GB4544.2
InternLM3 8B Instruct8.8B5.9 GB4538.7
Gemma 1 7B8.54B5.7 GB4624.7
CodeGemma 7B8.54B5.7 GB4640.2
LFM2 8B A1B8.3B5.6 GB26524.3
Seed-Coder 8B Instruct8.25B5.5 GB4834.1
Seed-Coder 8B Reasoning8.25B5.5 GB4832.9
DeepSeek R1-0528 Qwen3 8B8.2B5.5 GB4836.3
Qwen3-8B8.2B5.5 GB4843.3
Granite 3.0 8B8.17B5.5 GB4936.4
Granite 3.1 8B8.17B5.5 GB4938.6
Command-R7B8.03B5.4 GB4935.3
Aya Expanse 8B8B5.4 GB5027.8