FitMyLLM
▸ NVIDIA· TESLA

NVIDIA GeForce GT 230 OEM

Running LLMs on the GeForce GT 230 OEM — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
2 GB
BUDGET
BANDWIDTH
24
GB/S
MODELS Q4
25/449
6%
7B Q4 SPEED
~3
SLOW
▸ MODEL COVERAGE @ Q46% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
—
1.7GB NEEDED
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
2 GB
BANDWIDTH
24 GB/s
FP16 COMPUTE
0.2 TFLOPS
TDP
75W
MEMORY
DDR2
ARCHITECTURE
Tesla
CUDA CORES
96
PCIE
Gen 2 x16
22
FAST MODELS · >30 TOK/S
Real-time chat speed
25
USABLE · >10 TOK/S
Comfortable for all tasks
25
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test GeForce GT 230 OEM (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· 25
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
213
TOK/S · 36% VRAM
›
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
142
TOK/S · 38% VRAM
›
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
137
TOK/S · 38% VRAM
›
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
137
TOK/S · 38% VRAM
›
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
75
TOK/S · 43% VRAM
›
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
71
TOK/S · 44% VRAM
›
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
63
TOK/S · 45% VRAM
›
A
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
57
TOK/S · 46% VRAM
›
A
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
57
TOK/S · 46% VRAM
›
A
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
57
TOK/S · 46% VRAM
›
A
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
55
TOK/S · 47% VRAM
›
A
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
53
TOK/S · 47% VRAM
›
A
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
51
TOK/S · 48% VRAM
›
B
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
38
TOK/S · 53% VRAM
›
B
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
38
TOK/S · 53% VRAM
›
B
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
37
TOK/S · 54% VRAM
›
B
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
34
TOK/S · 56% VRAM
›
B
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
34
TOK/S · 56% VRAM
›
B
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
32
TOK/S · 57% VRAM
›
B
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
32
TOK/S · 57% VRAM
›
B
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
32
TOK/S · 57% VRAM
›
B
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
31
TOK/S · 58% VRAM
›
B
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
26
TOK/S · 63% VRAM
›
B
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
26
TOK/S · 63% VRAM
›
C
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
24
TOK/S · 66% VRAM
›
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your GeForce GT 230 OEM.

▸Find models for GeForce GT 230 OEMCompare with other GPUs
WHAT THIS CARD IS WORTH

GT 230 OEM holds 25 of the models in our catalogue and is, in practice, a Q4_K_M card — the largest it takes is Falcon-H1 1.5B 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
Falcon-H1 1.5B1.55B · Q4_K_M1.3 GB19 tok/sEST
Qwen2.5-Coder-1.5B1.5B · Q4_K_M1.3 GB20 tok/sEST
Qwen2 Math 1.5B1.5B · Q4_K_M1.3 GB20 tok/sEST
Qwen 2.5 1.5B1.5B · Q4_K_M1.3 GB20 tok/sEST
Stella en 1.5B v51.5B · Q4_K_M1.3 GB20 tok/sEST
EXAONE-4.0-1.2B1.3B · Q4_K_M1.3 GB23 tok/sEST
MiniCPM-V 4.61.3B · Q5_K_M1.3 GB19 tok/sEST
LFM2.5-1.2B-Thinking1.2B · Q5_K_M1.3 GB21 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—
GT 230 OEM2 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 GT 230 OEM — 2 GB VRAM.

▸ GEFORCE GT 230 OEM SPEC
BRAND
NVIDIA
VRAM
2 GB DDR2
BANDWIDTH
24 GB/s
FP16 COMPUTE
0.2 TFLOPS
FP32 COMPUTE
0.2 TFLOPS
CUDA CORES
96
TDP
75 W
ARCHITECTURE
Tesla
▸ AI CAPABILITY
25/ 449 models @ Q4

With 2 GB VRAM and 24 GB/s bandwidth, this GPU handles models up to 0.81B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~3 tok/s.

§ 01TOP MODELS FOR GEFORCE GT 230 OEM
25 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
GPT-2 Large 774M0.81B1.0 GB245.6
Qwen3 0.6B0.75B0.9 GB2619.1
LFM2 700M0.74B0.9 GB2650.4
Qwen 1.5 0.5B0.62B0.9 GB319.9
Qwen3-Embedding 0.6B0.6B0.9 GB32—
Falcon-H1R Tiny 0.6B0.6B0.9 GB32—
Falcon Perception 0.6B0.6B0.9 GB32—
BGE-M30.568B0.8 GB3463.0
Snowflake Arctic Embed L v2.00.568B0.8 GB34—
Falcon-H1 0.5B0.52B0.8 GB3741.7
Qwen 2.5 0.5B0.5B0.8 GB3819.4
SmolVLM 500M0.5B0.8 GB38—
GPT-2 Medium 345M0.38B0.7 GB515.9
SmolLM2 360M0.36B0.7 GB538.2
LFM2 350M0.35B0.7 GB5546.3
bge-large-en-v1.5 335M0.335B0.7 GB5762.3
mxbai-embed-large-v10.335B0.7 GB5764.7
Snowflake Arctic Embed L0.335B0.7 GB5756.0
Snowflake Arctic Embed M v2.00.305B0.7 GB63—
Gemma 3 270M0.27B0.7 GB7112.6