GaMS3 12B Instruct — Hardware Requirements & GPU Compatibility
Chat
GaMS3 12B Instruct is a 11.8B-parameter open language model from cjvt. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 8.11 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q3_K_M · 16K)Best on 24 GB (Q4_K_M · ~81 tok/s · 128K)Macs from 32 GB (~10 tok/s)
Gemma Terms
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run GaMS3 12B Instruct
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve cjvt/GaMS3-12B-Instruct --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 3060 12GB (3 GB headroom) · Q4_K_M
- Generation speed
- ~29 tok/s
- generation speed
- Cost per 1M output tokens
- $0.73
- per 1M output tokens
How Much VRAM Does GaMS3 12B Instruct Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 6.1 GB | 8.1 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~108 tok/s | |
| 6.8 GB | 8.9 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~97 tok/s | |
| 8.1 GB | 10.2 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~81 tok/s | |
| 9.4 GB | 11.5 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~69 tok/s | |
| 10.8 GB | 12.8 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~61 tok/s | |
| 12.8 GB | 14.9 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~51 tok/s | |
| 24.6 GB | 26.7 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 25.4 GB |
Selected Q4_K_M
4-bit medium quantization8.9GBat 16K context
- Weights
- 7.1 GB
- KV cache
- 1.6 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 15.1 GB free · ~81 tok/s
≥50 tok/s needs ≥623 GB/s · RTX 5090 ~144, RTX 3090 Ti ~81
Which Hardware Can Run GaMS3 12B Instruct?
Q4_K_M · 8.1 GBGaMS3 12B Instruct (Q4_K_M) requires 8.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 11+ GB is recommended. Using the full 131K context window can add up to 9.6 GB, bringing total usage to 17.7 GB. 40 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080 Ti. 155 devices with unified memory can run GaMS3 12B Instruct, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, iPad Pro M5 13" (16 GB).
12 GB
12 GB tier →fits · 3.1 GB free · up to 32K ctx
10 GPUs
RTX 3080 Ti~73Simulate output on RTX 3080 TiRTX 5070~54Simulate output on RTX 5070RTX 4070~40Simulate output on RTX 407020 GB
fits · 11.1 GB free · up to 128K ctx
1 GPU
RX 7900 XT~59Simulate output on RX 7900 XT32–40 GB
fits · 23.1 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~144Simulate output on RTX 5090M5 Max36 GB~40Simulate output on Mac Studio M5 Max (36 GB)M532 GB~13Simulate output on MacBook Air 13" M5 (32 GB)Orin~16Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~11Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
5| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Gemma 3 12B ITGoogle | 74.2 | 10.9 GB | ~82 | same size |
| Gemma 4 12BGoogle | — | 9 GB | ~80 | same size |
| Mistral Nemo Base 2407Mistral AI | — | 11 GB | ~81 | same size |
| Falcon 11BTII UAE | — | 8.5 GB | ~89 | same size |
| NVIDIA Nemotron Nano 12B v2NVIDIA | — | 11 GB | ~81 | same size |
Specifications
- Publisher
- cjvt
- Parameters
- 11.8B
- Architecture
- Gemma3ForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 262,208
- Release Date
- 2025-12-04
- License
- Gemma Terms
Frequently Asked Questions
How much VRAM does GaMS3 12B Instruct need?
GaMS3 12B Instruct requires 8.1 GB of VRAM at Q4_K_M, or 24.6 GB at BF16. Full 131K context adds up to 9.6 GB (17.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 11.8B × 4.8 bits ÷ 8 = 7.1 GB
KV Cache + Overhead ≈ 1 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 10.6 GB (at full 131K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run GaMS3 12B Instruct?
Yes, at Q8_0 (12.8 GB) or lower. Higher quantizations like BF16 (24.6 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for GaMS3 12B Instruct?
For GaMS3 12B Instruct, Q4_K_M (8.1 GB) offers the best balance of quality and VRAM usage. Q5_K_M (9.4 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 6.1 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run GaMS3 12B Instruct on a Mac?
Yes, but only at lower quantizations. The smallest Mac that can run GaMS3 12B Instruct is Apple iPhone 17 Pro (12 GB) at Q2_K; 139 of the 156 Macs we list can run it at some quantization. For Q4_K_M (8.1 GB) you need a Mac with more unified memory.
Can I run GaMS3 12B Instruct locally?
Yes — GaMS3 12B Instruct can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is GaMS3 12B Instruct?
At Q4_K_M, GaMS3 12B Instruct can reach ~592 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~81 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 8.11 × 0.65 = ~641 tok/s
Estimated speed at Q4_K_M (8.1 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of GaMS3 12B Instruct?
At Q4_K_M, the download is about 7.06 GB. The full-precision BF16 version is 23.53 GB. The smallest option (Q2_K) is 5.00 GB.
Which GPUs can run GaMS3 12B Instruct?
40 consumer GPUs can run GaMS3 12B Instruct at Q4_K_M. Top options include AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 6900 XT, AMD Radeon RX 6700 XT. 26 GPUs have plenty of headroom for comfortable inference.
Which devices can run GaMS3 12B Instruct?
155 devices with unified memory can run GaMS3 12B Instruct at Q4_K_M (8.1 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.