FitMyLLM
NVIDIA· Ada Lovelace

NVIDIA GeForce RTX 4080 Max-Q

Running LLMs on the GeForce RTX 4080 Max-Q — the long read: which models fit at which quantisation, and the settings worth changing. · Or what else $1,199 buys

VRAM
12 GB
Entry-level
Bandwidth
432
GB/s
Models Q4
263/498
53%
7B Q4 speed
~49
Fast
Model coverage at Q453% of all
Estimated speed· By model size at Q4

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

3B
~115
tok/s
7B
~49
tok/s
14B
~25
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
12 GB
Bandwidth
432 GB/s
FP16 Compute
20 TFLOPS
TDP
60W
Memory
GDDR6
Architecture
Ada Lovelace
CUDA Cores
7,424
Tensor Cores
232
PCIe
Gen 4 x16
MSRP
$1,199
237
Fast models · >30 tok/s
Real-time chat speed
263
Usable · >10 tok/s
Comfortable for all tasks
263
Total compatible
Fit in VRAM at Q4
Where to buy

MSRP $1,199

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

Don’t want to buy?

Test GeForce RTX 4080 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· 263
S
Falcon-H1R Tiny 90M0.09B
Falcon·256K ctx· chat· reasoning
3840
tok/s · 5% VRAM
›
S
SmolLM2 135M0.135B
Smollm·2K ctx· chat
2560
tok/s · 5% VRAM
›
S
nomic-embed-text-v1.5 100M0.14B
Embedding·8K ctx· chat
2469
tok/s · 5% VRAM
›
S
GPT-2 124M0.14B
Gpt2·1K ctx· chat
2469
tok/s · 5% VRAM
›
S
SmolVLM 256M0.256B
Smollm·8K ctx· chat· vision
1350
tok/s · 5% VRAM
›
S
Gemma 3 270M0.27B
Gemma·32K ctx· chat
1280
tok/s · 5% VRAM
›
S
Snowflake Arctic Embed M v2.00.305B
Embedding·8K ctx· embedding· multilingual
1133
tok/s · 6% VRAM
›
S
bge-large-en-v1.5 335M0.335B
Embedding·1K ctx· chat
1032
tok/s · 6% VRAM
›
S
mxbai-embed-large-v10.335B
Embedding·1K ctx· embedding
1032
tok/s · 6% VRAM
›
S
Snowflake Arctic Embed L0.335B
Embedding·1K ctx· embedding
1032
tok/s · 6% VRAM
›
S
LFM2 350M0.35B
Lfm·125K ctx· chat· tool_use
987
tok/s · 6% VRAM
›
S
SmolLM2 360M0.36B
Smollm·8K ctx· chat
960
tok/s · 6% VRAM
›
S
GPT-2 Medium 345M0.38B
Gpt2·1K ctx· chat
909
tok/s · 6% VRAM
›
S
Qwen 2.5 0.5B0.5B
Qwen·32K ctx· chat
691
tok/s · 7% VRAM
›
S
SmolVLM 500M0.5B
Smollm·8K ctx· chat· vision
691
tok/s · 7% VRAM
›
S
Falcon-H1 0.5B0.52B
Falcon·128K ctx· chat
665
tok/s · 7% VRAM
›
S
BGE-M30.568B
Embedding·8K ctx· embedding
608
tok/s · 7% VRAM
›
S
Snowflake Arctic Embed L v2.00.568B
Embedding·8K ctx· embedding· multilingual
608
tok/s · 7% VRAM
›
S
Qwen3-Embedding 0.6B0.6B
Embedding·32K ctx· embedding· multilingual
576
tok/s · 7% VRAM
›
S
Falcon-H1R Tiny 0.6B0.6B
Falcon·32K ctx· chat· reasoning
576
tok/s · 7% VRAM
›
S
Falcon Perception 0.6B0.6B
Falcon·4K ctx· vision
576
tok/s · 7% VRAM
›
S
Qwen 1.5 0.5B0.62B
Qwen·32K ctx· chat
557
tok/s · 7% VRAM
›
S
LFM2 700M0.74B
Lfm·125K ctx· chat· tool_use
467
tok/s · 8% VRAM
›
S
Qwen3 0.6B0.75B
Qwen·32K ctx· chat· reasoning
461
tok/s · 8% VRAM
›
S
GPT-2 Large 774M0.81B
Gpt2·1K ctx· chat
427
tok/s · 8% VRAM
›
S
Qwen 3.5 0.8B0.87B
Qwen·256K ctx· chat· coding· multilingual
397
tok/s · 9% VRAM
›
S
InternVL3 1B0.94B
Other·32K ctx· chat· vision
368
tok/s · 9% VRAM
›
S
MiniCPM5 1B1.08B
Minicpm·128K ctx· chat· reasoning· multilingual
320
tok/s · 10% VRAM
›
S
K2 Horizon 0.9B1.08B
Other·128K ctx· chat· coding· reasoning
320
tok/s · 10% VRAM
›
S
TinyLlama 1.1B1.1B
Llama·2K ctx· chat
314
tok/s · 10% VRAM
›
S
LFM2.5-1.2B-Thinking1.2B
Lfm·122K ctx· chat· reasoning· tool_use
288
tok/s · 10% VRAM
›
S
Llama-3.2-1B1.2B
Llama·4K ctx· chat
288
tok/s · 10% VRAM
›
S
LFM2 1.2B1.2B
Lfm·125K ctx· chat· tool_use· multilingual
288
tok/s · 10% VRAM
›
S
Zamba2 1.2B1.2B
Other·4K ctx· chat
288
tok/s · 10% VRAM
›
S
Ling 3.0 Tiny7.89BMoE
Other·128K ctx· chat· coding· reasoning
266
tok/s · 44% VRAM
›
S
EXAONE-4.0-1.2B1.3B
Exaone·64K ctx· chat
266
tok/s · 11% VRAM
›
S
OPT 1.3B1.3B
Opt·2K ctx· chat
266
tok/s · 11% VRAM
›
S
MiniCPM-V 4.61.3B
Other·256K ctx· chat· vision
266
tok/s · 11% VRAM
›
S
DeepSeek Coder 1.3B1.35B
Deepseek·16K ctx· coding
256
tok/s · 11% VRAM
›
S
Phi-1 1.3B1.42B
Phi·2K ctx· coding
243
tok/s · 11% VRAM
›
Next step

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your GeForce RTX 4080 Max-Q.

What this card is worth

RTX 4080 Max-Q holds 263 of the models in our catalogue and is, in practice, a IQ4_XS card — the largest it takes is Ling-lite 16.8B at IQ4_XS.

Tokens/sec per $100
4.0
8B at Q4_K_M, so cards compare like for like
VRAM per $100
1.00 GB
what memory costs on this card
The biggest it takes
Ling-lite 16.8B16.8B · IQ4_XS10.7 GB162 tok/sest
DeepSeek V2 Lite 16B16B · Q4_K_S10.6 GB155 tok/sest
StarCoder2 15B15.96B · Q4_K_S10.7 GB23 tok/sest
DeepSeek-Coder-V2-Lite 15.7B15.7B · Q4_K_S10.4 GB155 tok/sest
DeepSeek R1 Distill Qwen 14B14.8B · Q4_K_S10.4 GB25 tok/sest
DeepCoder 14B14.8B · Q4_K_S10.4 GB25 tok/sest
Qwen2.5-Coder-14B14.8B · Q4_K_S10.4 GB25 tok/sest
Qwen2.5-14B14.8B · Q4_K_S10.4 GB25 tok/sest
Against cards of similar memory
M4 Pro (24GB)16 GB$1,3992.5 tok/s per $100
M4 (24GB)16 GB$6992.9 tok/s per $100
M2 (24GB)16 GB$9991.7 tok/s per $100
M3 (24GB)16 GB$9991.7 tok/s per $100
RTX 4080 Max-Q12 GB$1,1994.0 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 GeForce RTX 4080 Max-Q — 12 GB VRAM.

GeForce RTX 4080 Max-Q spec
Brand
NVIDIA
VRAM
12 GB GDDR6
Bandwidth
432 GB/s
FP16 Compute
20 TFLOPS
FP32 Compute
20 TFLOPS
CUDA Cores
7,424
Tensor Cores
232
TDP
60 W
Architecture
Ada Lovelace
MSRP
$1199
AI capability
263/ 498 models @ Q4

With 12 GB VRAM and 432 GB/s bandwidth, this GPU handles models up to 14.8B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~49 tok/s.

Rent it — live prices
Checking what the clouds are charging…
Top models for GeForce RTX 4080 Max-Q
263 fit · showing 20
ModelSizeVRAM Q4tok/sAvg
DeepSeek R1 Distill Qwen 14B14.8B9.5 GB2343.9
DeepCoder 14B14.8B9.5 GB2336.0
Qwen2.5-Coder-14B14.8B9.5 GB2341.3
Qwen2.5-14B14.8B9.5 GB2341.3
Qwen3 14B14.8B9.5 GB2345.7
phi-4 14B14.66B9.4 GB2433.7
Phi-4-reasoning 14B14.66B9.4 GB2433.7
Phi-4-reasoning-plus 14B14.66B9.4 GB2472.2
Phi-3-medium-14b14B9.0 GB2533.7
Qwen 1.5 14B14B9.0 GB2541.3
Ministral 3 14B Reasoning13.95B9.0 GB2573.6
Baichuan2 13B13B8.4 GB2723.6
Llama 2 13B13B8.4 GB2717.2
CodeLlama 13B13B8.4 GB2719.7
Vicuna 13B13B8.4 GB2711.8
LLaMA 1 13B13B8.4 GB2732.9
OPT 13B13B8.4 GB2735.8
Orca 2 13B13B8.4 GB2725.4
WizardCoder Python 13B13B8.4 GB2760.1
WizardLM 13B13B8.4 GB2719.5