Can I run Llama-3.1-8B-Instruct on an RTX 3090 (24GB)?
Computed with the open FitLLM engine — accurate per-layer KV-cache modeling, not a naive estimate. Updated 2026-07-10.
Memory breakdown (Q4_K_M, F16 KV, 33K context)
| Model weights | 4.6 GB |
| KV cache | 4.0 GB |
| Runtime overhead + reserve | 4.1 GB |
| Total used | 12.7 / 24 GB |
| Free | 11.3 GB |
Max context that fits at Q4_K_M: ~92K tokens · with Q8 KV cache → ~131K tokens.
Every quantization on the RTX 3090
| Weight quant | Weights | Fits (KV F16) | Used @32K |
|---|---|---|---|
| Q4_K_M | 4.6 GB | ✅ up to 92K ctx | 12.7 / 24.0 GB |
| Q5_K_M | 5.3 GB | ✅ up to 87K ctx | 13.5 / 24.0 GB |
| Q6_K | 6.1 GB | ✅ up to 82K ctx | 14.4 / 24.0 GB |
| Q8_0 | 7.9 GB | ✅ up to 70K ctx | 16.5 / 24.0 GB |
| FP16 | 14.9 GB | ❌ up to 24K ctx | 24.3 / 24.0 GB |
Lower weight quants free memory at some output-quality cost — Q4 is the common sweet spot; below that quality drops faster.
KV cache is F16 here (llama.cpp default). Drop it to Q8/Q4 (-ctk/-ctv) for more context.
Embed this verdict
Live badge for your README or model card — recomputed by the engine, never stale:
[](https://fitllm.run/can-i-run/llama-3-1-8b-instruct-on-rtx-3090)
Or from your terminal (exit 0/1 — works as a pre-download guard):
npx fitllm "Llama-3.1-8B-Instruct" --gpu "RTX 3090"Why most VRAM calculators get this wrong
Llama-3.1-8B-Instruct's KV cache is computed per layer with its real head_dim and grouped-query head count — not the uniform "all layers × full context" shortcut most calculators use.
Other options
same GPU Models that fit on the RTX 3090: GLM-4.7-Flash, gpt-oss-20b, Qwen 3.6 27B, Gemma 4 e2b, Gemma 4 e4b, Gemma 4 12b, Gemma 4 26b A4B, Llama-3.2-3B-Instruct, Llama-3.1-8B-Instruct, Qwen3-0.6B, Qwen3-1.7B, Llama-3.2-1B-Instruct, Gemma-3-1B-it.
same model GPUs that run Llama-3.1-8B-Instruct: RTX 5090 (32GB), RTX 5080 (16GB), RTX 5070 Ti (16GB), RTX 5070 (12GB), RTX 4090 (24GB), RTX 4080 SUPER (16GB), RTX 4070 Ti SUPER (16GB), RTX 4070 (12GB), RTX 4060 Ti 16GB (16GB), RTX 3090 (24GB), RTX 3090 Ti (24GB), RTX 3080 10GB (10GB), RTX 3080 12GB (12GB), RTX 3060 12GB (12GB), RTX 5060 Ti 16GB (16GB), RTX 4070 Ti (12GB), RTX 4080 (16GB), RTX 2080 Ti (11GB), RTX 6000 Ada (48GB), RTX PRO 6000 Blackwell (96GB), RX 7900 XTX (24GB), RX 7900 XT (20GB), RX 7800 XT (16GB), RX 9070 XT (16GB), RX 9070 (16GB), Radeon PRO W7900 (48GB), 2× RTX 3090 (48GB), 2× RTX 4090 (48GB), 4× RTX 3090 (96GB), A100 40GB (40GB), A100 80GB (80GB), H100 80GB (80GB), H200 141GB (141GB), B200 (192GB).
Reproduce it
Llama-3.1-8B-Instruct = 8B, 32 layers. The RTX 3090 has 24GB / 936GB/s. Same math, open source: fitllm-engine. GGUF bpw from llama.cpp.
All numbers are computed by the open-source fitllm-engine (MIT) from official model config.json values — reproduce or audit them yourself. Estimates; real usage varies with runtime (llama.cpp / MLX / Ollama), driver and display. Found a mismatch? Report it. · FitLLM home