FitMyLLM
NVIDIA· KEPLER

NVIDIA GeForce GTX 780 Ti

Running LLMs on the GeForce GTX 780 Ti — the long read: which models fit at which quantisation, and the settings worth changing. · Or what a budget buys

VRAM
3 GB
BUDGET
BANDWIDTH
337
GB/S
MODELS Q4
81/449
18%
7B Q4 SPEED
~39
FAST
▸ MODEL COVERAGE @ Q418% OF ALL
▸ ESTIMATED SPEED· BY MODEL SIZE @ Q4

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

3B
~90
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
3 GB
BANDWIDTH
337 GB/s
FP16 COMPUTE
5.3 TFLOPS
TDP
250W
MEMORY
GDDR5
ARCHITECTURE
Kepler
CUDA CORES
2,880
PCIE
Gen 3 x16
81
FAST MODELS · >30 TOK/S
Real-time chat speed
81
USABLE · >10 TOK/S
Comfortable for all tasks
81
TOTAL COMPATIBLE
Fit in VRAM at Q4
▸ DON’T WANT TO BUY?

Test GeForce GTX 780 Ti (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· 81
S
Falcon-H1R Tiny 90M0.09B
FALCON·256K CTX· CHAT· REASONING
2996
TOK/S · 18% VRAM
S
SmolLM2 135M0.135B
SMOLLM·2K CTX· CHAT
1997
TOK/S · 19% VRAM
S
nomic-embed-text-v1.5 100M0.14B
EMBEDDING·8K CTX· CHAT
1926
TOK/S · 19% VRAM
S
GPT-2 124M0.14B
GPT2·1K CTX· CHAT
1926
TOK/S · 19% VRAM
S
SmolVLM 256M0.256B
SMOLLM·8K CTX· CHAT· VISION
1053
TOK/S · 21% VRAM
S
Gemma 3 270M0.27B
GEMMA·32K CTX· CHAT
999
TOK/S · 22% VRAM
S
Snowflake Arctic Embed M v2.00.305B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
884
TOK/S · 22% VRAM
S
bge-large-en-v1.5 335M0.335B
EMBEDDING·1K CTX· CHAT
805
TOK/S · 23% VRAM
S
mxbai-embed-large-v10.335B
EMBEDDING·1K CTX· EMBEDDING
805
TOK/S · 23% VRAM
S
Snowflake Arctic Embed L0.335B
EMBEDDING·1K CTX· EMBEDDING
805
TOK/S · 23% VRAM
S
LFM2 350M0.35B
LFM·125K CTX· CHAT· TOOL_USE
770
TOK/S · 23% VRAM
S
SmolLM2 360M0.36B
SMOLLM·8K CTX· CHAT
749
TOK/S · 24% VRAM
S
GPT-2 Medium 345M0.38B
GPT2·1K CTX· CHAT
709
TOK/S · 24% VRAM
S
Qwen 2.5 0.5B0.5B
QWEN·32K CTX· CHAT
539
TOK/S · 26% VRAM
S
SmolVLM 500M0.5B
SMOLLM·8K CTX· CHAT· VISION
539
TOK/S · 26% VRAM
S
Falcon-H1 0.5B0.52B
FALCON·128K CTX· CHAT
518
TOK/S · 27% VRAM
S
BGE-M30.568B
EMBEDDING·8K CTX· EMBEDDING
475
TOK/S · 28% VRAM
S
Snowflake Arctic Embed L v2.00.568B
EMBEDDING·8K CTX· EMBEDDING· MULTILINGUAL
475
TOK/S · 28% VRAM
S
Qwen3-Embedding 0.6B0.6B
EMBEDDING·32K CTX· EMBEDDING· MULTILINGUAL
449
TOK/S · 29% VRAM
S
Falcon-H1R Tiny 0.6B0.6B
FALCON·32K CTX· CHAT· REASONING
449
TOK/S · 29% VRAM
S
Falcon Perception 0.6B0.6B
FALCON·4K CTX· VISION
449
TOK/S · 29% VRAM
S
Qwen 1.5 0.5B0.62B
QWEN·32K CTX· CHAT
435
TOK/S · 29% VRAM
S
LFM2 700M0.74B
LFM·125K CTX· CHAT· TOOL_USE
364
TOK/S · 31% VRAM
S
Qwen3 0.6B0.75B
QWEN·32K CTX· CHAT· REASONING
359
TOK/S · 32% VRAM
S
GPT-2 Large 774M0.81B
GPT2·1K CTX· CHAT
333
TOK/S · 33% VRAM
S
Qwen 3.5 0.8B0.87B
QWEN·256K CTX· CHAT· CODING· MULTILINGUAL
310
TOK/S · 34% VRAM
S
InternVL3 1B0.94B
OTHER·32K CTX· CHAT· VISION
287
TOK/S · 35% VRAM
S
MiniCPM5 1B1.08B
MINICPM·128K CTX· CHAT· REASONING· MULTILINGUAL
250
TOK/S · 38% VRAM
S
TinyLlama 1.1B1.1B
LLAMA·2K CTX· CHAT
245
TOK/S · 39% VRAM
S
LFM2.5-1.2B-Thinking1.2B
LFM·122K CTX· CHAT· REASONING· TOOL_USE
225
TOK/S · 41% VRAM
S
Llama-3.2-1B1.2B
LLAMA·4K CTX· CHAT
225
TOK/S · 41% VRAM
S
LFM2 1.2B1.2B
LFM·125K CTX· CHAT· TOOL_USE· MULTILINGUAL
225
TOK/S · 41% VRAM
S
Zamba2 1.2B1.2B
OTHER·4K CTX· CHAT
225
TOK/S · 41% VRAM
S
EXAONE-4.0-1.2B1.3B
EXAONE·64K CTX· CHAT
207
TOK/S · 43% VRAM
S
OPT 1.3B1.3B
OPT·2K CTX· CHAT
207
TOK/S · 43% VRAM
S
MiniCPM-V 4.61.3B
OTHER·256K CTX· CHAT· VISION
207
TOK/S · 43% VRAM
S
DeepSeek Coder 1.3B1.35B
DEEPSEEK·16K CTX· CODING
200
TOK/S · 44% VRAM
S
Phi-1 1.3B1.42B
PHI·2K CTX· CODING
190
TOK/S · 45% VRAM
S
Phi-1.5 1.3B1.42B
PHI·2K CTX· CHAT· CODING
190
TOK/S · 45% VRAM
S
Qwen2.5-Coder-1.5B1.5B
QWEN·32K CTX· CHAT· TOOL_USE· CODING
180
TOK/S · 47% VRAM
▸ NEXT STEP

Get personalized recommendations.

See ranked models with benchmark scores, run commands, and precise speed estimates for your GeForce GTX 780 Ti.

WHAT THIS CARD IS WORTH

GTX 780 Ti holds 81 of the models in our catalogue and is, in practice, a Q4_K_M card — the largest it takes is Falcon3-3B 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
Falcon3-3B3.23B · Q4_K_M2.6 GB91 tok/sEST
granite-4.0-h-micro 3.2B3.2B · Q4_K_M2.6 GB91 tok/sEST
Llama-3.2-3B3.2B · Q4_K_M2.7 GB91 tok/sEST
Falcon-H1 3B3.15B · Q5_K_M2.7 GB80 tok/sEST
Qwen 2.5 3B3.1B · Q5_K_M2.6 GB81 tok/sEST
SmolLM3-3B3.1B · Q4_K_M2.5 GB94 tok/sEST
Ministral 3B3B · Q4_K_M2.6 GB98 tok/sEST
StarCoder2 3B3B · Q5_K_M2.5 GB84 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
RTX 2060 6GB6 GB$15026.7 tok/s per $100
GTX 780 Ti3 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 780 Ti — 3 GB VRAM.

GEFORCE GTX 780 TI SPEC
BRAND
NVIDIA
VRAM
3 GB GDDR5
BANDWIDTH
337 GB/s
FP16 COMPUTE
5.3 TFLOPS
FP32 COMPUTE
5.3 TFLOPS
CUDA CORES
2,880
TDP
250 W
ARCHITECTURE
Kepler
▸ AI CAPABILITY
81/ 449 models @ Q4

With 3 GB VRAM and 337 GB/s bandwidth, this GPU handles models up to 3.1B parameters.

Speed ≈ bandwidth / model_size × efficiency. A 7B model at Q4 runs at ~39 tok/s.

§ 01TOP MODELS FOR GEFORCE GTX 780 TI
81 FIT · SHOWING 20
MODELSIZEVRAM Q4TOK/SAVG
Qwen 2.5 3B3.1B2.4 GB8737.2
SmolLM3-3B3.1B2.4 GB8730.5
Ministral 3B3B2.3 GB9029.6
StarCoder2 3B3B2.3 GB909.5
Granite 4.1 3B3B2.3 GB9016.6
xLAM-2 3B Function-Calling3B2.3 GB90
Jamba 2 3B3B2.3 GB90
Dolly v2 3B2.8B2.2 GB965.6
StableLM Zephyr 3B2.79B2.2 GB9714.9
Zephyr 3B2.79B2.2 GB9714.4
OPT 2.7B2.7B2.1 GB10028.0
Phi-2 2.7B2.7B2.1 GB10024.1
Zamba2 2.7B2.7B2.1 GB10048.0
Granite 3.0 2B2.63B2.1 GB10335.8
gemma-2-2b2.6B2.1 GB10422.9
LFM2 2.6B2.6B2.1 GB10416.3
Granite 3.1 2B2.53B2.0 GB10737.8
Granite 3.3 2B2.53B2.0 GB10720.5
Gemma 1 2B2.51B2.0 GB10720.2
CodeGemma 2B2.51B2.0 GB10722.9