▸ NVIDIA· FERMI

NVIDIA GeForce GTX 470

Running LLMs on the GeForce GTX 470 — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM

1 GB

BUDGET

BANDWIDTH

133.9

GB/S

MODELS Q4

16/449

7B Q4 SPEED

~15

GOOD

▸ MODEL COVERAGE @ Q44% OF ALL

▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

—

1.7GB NEEDED

—

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

1 GB

BANDWIDTH

133.9 GB/s

FP16 COMPUTE

1.1 TFLOPS

TDP

215W

MEMORY

GDDR5

ARCHITECTURE

Fermi

CUDA CORES

448

PCIE

Gen 2 x16

FAST MODELS · >30 TOK/S

Real-time chat speed

USABLE · >10 TOK/S

Comfortable for all tasks

TOTAL COMPATIBLE

Fit in VRAM at Q4

▸ DON’T WANT TO BUY?

Test GeForce GTX 470 (or anything bigger) without committing. Pay by the second, cancel anytime.

Rent on Vast.ai →Or RunPod →

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· 16

Falcon-H1R Tiny 90M0.09B

FALCON·256K CTX· CHAT· REASONING

nomic-embed-text-v1.5 100M0.14B

EMBEDDING·8K CTX· CHAT

SMOLLM·8K CTX· CHAT· VISION

Snowflake Arctic Embed M v2.00.305B

EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL

bge-large-en-v1.5 335M0.335B

EMBEDDING·1K CTX· CHAT

mxbai-embed-large-v10.335B

EMBEDDING·1K CTX· EMBEDDING

Snowflake Arctic Embed L0.335B

EMBEDDING·1K CTX· EMBEDDING

LFM·125K CTX· CHAT· TOOL_USE

GPT-2 Medium 345M0.38B

SMOLLM·8K CTX· CHAT· VISION

FALCON·128K CTX· CHAT

▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your GeForce GTX 470.

▸Find models for GeForce GTX 470 Compare with other GPUs

WHAT THIS CARD IS WORTH

GTX 470 holds 16 of the models in our catalogue and is, in practice, a Q4_K_M card — the largest it takes is TinyLlama 1.1B 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

TinyLlama 1.1B1.1B · Q4_K_M1.1 GB134 tok/sEST

MiniCPM5 1B1.08B · Q4_K_M1.0 GB136 tok/sEST

InternVL3 1B0.94B · Q5_K_M1.0 GB134 tok/sEST

Qwen 3.5 0.8B0.87B · Q6_K1.1 GB126 tok/sEST

LFM2 700M0.74B · Q6_K1.0 GB148 tok/sEST

Qwen 1.5 0.5B0.62B · Q5_K_M1.1 GB204 tok/sEST

Qwen3-Embedding 0.6B0.6B · Q4_K_M1.1 GB245 tok/sEST

Falcon-H1R Tiny 0.6B0.6B · Q8_01.0 GB141 tok/sEST

AGAINST CARDS OF SIMILAR MEMORY

M2 (8GB)5 GB$5992.8 tok/s per $100

M3 (8GB)5 GB$5992.8 tok/s per $100

M1 (8GB)5 GB$4992.2 tok/s per $100

Radeon RX 6600S4 GB$175—

GTX 4701 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 GeForce GTX 470 — 1 GB VRAM.

▸ GEFORCE GTX 470 SPEC

BRAND: NVIDIA
VRAM: 1 GB GDDR5
BANDWIDTH: 133.9 GB/s
FP16 COMPUTE: 1.1 TFLOPS
FP32 COMPUTE: 1.1 TFLOPS
CUDA CORES: 448
TDP: 215 W
ARCHITECTURE: Fermi

▸ AI CAPABILITY

16/ 449 models @ Q4

With 1 GB VRAM and 133.9 GB/s bandwidth, this GPU handles models up to 0.52B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~15 tok/s.

§ 01TOP MODELS FOR GEFORCE GTX 470

16 FIT · SHOWING 16

MODEL	SIZE	VRAM Q4	TOK/S	AVG
Falcon-H1 0.5B	0.52B	0.8 GB	206	41.7
Qwen 2.5 0.5B	0.5B	0.8 GB	214	19.4
SmolVLM 500M	0.5B	0.8 GB	214	—
GPT-2 Medium 345M	0.38B	0.7 GB	282	5.9
SmolLM2 360M	0.36B	0.7 GB	298	8.2
LFM2 350M	0.35B	0.7 GB	306	46.3
bge-large-en-v1.5 335M	0.335B	0.7 GB	320	62.3
mxbai-embed-large-v1	0.335B	0.7 GB	320	64.7
Snowflake Arctic Embed L	0.335B	0.7 GB	320	56.0
Snowflake Arctic Embed M v2.0	0.305B	0.7 GB	351	—
Gemma 3 270M	0.27B	0.7 GB	397	12.6
SmolVLM 256M	0.256B	0.6 GB	418	28.3
nomic-embed-text-v1.5 100M	0.14B	0.6 GB	765	62.3
GPT-2 124M	0.14B	0.6 GB	765	6.5
SmolLM2 135M	0.135B	0.6 GB	793	7.0
Falcon-H1R Tiny 90M	0.09B	0.5 GB	1190	—