Home / Models / Gemma 3 12B Instruct
Gemma 3 12B Instruct
12B parameters · native context 131,072 · gemma3
Estimated VRAM requirement
Weight memory uses parameter count × bits per weight, plus a quantization overhead factor. KV cache uses standard GQA formulas. Values are calculated estimates, not measured allocations.
| Quantization | Estimated weights | 2K total | 4K total | 8K total | 16K total | 32K total | 64K total | 128K total |
|---|---|---|---|---|---|---|---|---|
| FP16 | 23.2 GB | 26.5 GB | 27.3 GB | 28.8 GB | 31.8 GB | 37.8 GB | 49.8 GB | 73.8 GB |
| Q8 | 13.0 GB | 15.6 GB | 16.3 GB | 17.8 GB | 20.8 GB | 26.8 GB | 38.8 GB | 62.8 GB |
| Q6 | 10.1 GB | 12.5 GB | 13.2 GB | 14.7 GB | 17.7 GB | 23.7 GB | 35.7 GB | 59.7 GB |
| Q5 | 8.6 GB | 10.8 GB | 11.5 GB | 13.0 GB | 16.0 GB | 22.0 GB | 34.0 GB | 58.0 GB |
| Q4 | 7.2 GB | 9.2 GB | 10.0 GB | 11.5 GB | 14.5 GB | 20.5 GB | 32.5 GB | 56.5 GB |
| Q3 | 5.7 GB | 7.7 GB | 8.4 GB | 9.9 GB | 12.9 GB | 18.9 GB | 30.9 GB | 54.9 GB |
Find a GPU for this model
Run at the quantization and context selected below. Defaults are Q4 and 8K when you do not change the controls. Results are full-VRAM fits only — not a speed ranking.
- Run at
- Q4 · 8K context
- Required VRAM
- 11.5 GB
Current cheapest full-VRAM option at Q4 · 8K: NVIDIA Quadro P5000 16 GB $330.99. See all value options · Build a PC for this model · Compatible workstations · Rent a server · Buy vs rent.
Lowest-cost full-VRAM option
Sorted by lowest current Amazon price among GPUs that fit entirely in VRAM. This is not a performance ranking.
NVIDIA Quadro P5000 16 GB
- GPU VRAM
- 16 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 14.4 GB
- VRAM headroom
- 2.9 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 5 mapped · 3 Amazon-matched
- Amazon
- $330.99 current Amazon price
NVIDIA RTX 5000 16 GB
- GPU VRAM
- 16 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 14.4 GB
- VRAM headroom
- 2.9 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 2 mapped · 1 Amazon-matched
- Amazon
- $596.99 current Amazon price
AMD Radeon RX 7600 XT 16 GB
- GPU VRAM
- 16 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 14.4 GB
- VRAM headroom
- 2.9 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 3 mapped · 1 Amazon-matched
- Amazon
- $599.00 current Amazon price
More VRAM headroom
Sorted by leftover usable VRAM after the calculated requirement. Headroom is not tokens/sec.
NVIDIA B200 192 GB
- GPU VRAM
- 192 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 172.8 GB
- VRAM headroom
- 161.3 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA H200 141 GB
- GPU VRAM
- 141 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 126.9 GB
- VRAM headroom
- 115.4 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA GH200 96 GB
- GPU VRAM
- 96 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 86.4 GB
- VRAM headroom
- 74.9 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
Other compatible GPUs
NVIDIA RTX 4500 24 GB
- GPU VRAM
- 24 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 21.6 GB
- VRAM headroom
- 10.1 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 2 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA RTX 3090 24 GB
- GPU VRAM
- 24 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 21.6 GB
- VRAM headroom
- 10.1 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 18 mapped · 9 Amazon-matched
- Amazon
- $1347.22 current Amazon price
NVIDIA RTX 5000 32 GB
- GPU VRAM
- 32 GB advertised
- Required VRAM
- 11.5 GB
- Usable GPU VRAM
- 28.8 GB
- VRAM headroom
- 17.3 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 2 mapped · 1 Amazon-matched
- Amazon
- $4479.96 current Amazon price
Compatible GPUs
Status is a calculated memory fit against usable VRAM (advertised × 0.9). Open a GPU to see product SKUs and Amazon listings.
Compare GPUs for this model
Published chip comparisons where both GPUs fully fit this model at Q4 / 8K. Opens with this model preselected; that query is not indexed.
Sources
Model specifications: official repository google/gemma-3-12b-it · revision official-fallback-config. Compatibility: RigForAI calculator v1.0.0 (estimated VRAM). GPU products: Icecat. Amazon is a separate affiliate match and is not used as a spec source.