Can an RTX 3090 run Gemma 3 12B?
Yes: an RTX 3090 runs Gemma 3 12B at 78.7 tok/s at Q4_K_M, measured on ollama with a 118-token prompt in October 2026.
1 signed run on llm-speed. Single-request decode speed in tokens per second; the short-chat workload unless noted. This run was measured by llm-speed (@llmspeed). Runs measured by llm-speed are credited to @llmspeed.
Measured tok/s by quant and runtime
| Quant | Runtime | Build | Decode tok/s (best run) | Workload | Runs | Source |
|---|---|---|---|---|---|---|
| Q4_K_M | ollama | 0.31.1 | 78.7 | chat-short | 1 | Measured by llm-speed@llmspeed |
Fastest quant that fits: Q4_K_M on ollama, 78.7 tok/s (measured by llm-speed; it fits because it ran on this hardware). Quants without a run are not listed.
Every measured row
| Run | Date (UTC) | Workload | Quant | Runtime / build | Prompt tokens | Output tokens | Decode tok/s | Prefill tok/s | TTFT ms | VRAM GiB | Source |
|---|---|---|---|---|---|---|---|---|---|---|---|
| r_3gnxkzhhmyj | 2026-10-16 | chat-short | Q4_K_M | ollama · 0.31.1 | 118 | 256 | 78.7 | 26.0 | 4534 | – | Measured by llm-speed@llmspeed |
| r_3gnxkzhhmyj | 2026-10-16 | chat-long | Q4_K_M | ollama · 0.31.1 | 3185 | 751 | 74.2 | 1648.1 | 1933 | – | Measured by llm-speed@llmspeed |
| r_3gnxkzhhmyj | 2026-10-16 | agent-trace | Q4_K_M | ollama · 0.31.1 | 3934 | 840 | 74.5 | 2887.7 | 808 | – | Measured by llm-speed@llmspeed |
| r_3gnxkzhhmyj | 2026-10-16 | concurrent-decode | Q4_K_M | ollama · 0.31.1 | 1024 | 256 | 76.0 | – | – | – | Measured by llm-speed@llmspeed |
Questions
Can an RTX 3090 run Gemma 3 12B?
Yes: an RTX 3090 runs Gemma 3 12B at 78.7 tok/s at Q4_K_M, measured on ollama with a 118-token prompt in October 2026.
How many tokens per second does Gemma 3 12B get on an RTX 3090?
Decode speed measured by llm-speed, by quantization and runtime: Q4_K_M on ollama: 78.7 tok/s. Based on 1 run measured by llm-speed; single-request chat-short unless noted.
What is the fastest quant of Gemma 3 12B that fits on an RTX 3090?
Q4_K_M on ollama was the fastest measured setting at 78.7 tok/s; it fit because it ran. Other quants are listed only where a run measured them.
How much VRAM does Gemma 3 12B use on an RTX 3090?
Memory use was not recorded for these runs. The VRAM calculator gives an estimate, not a measurement.