# GeForce RTX 5090 32 GB

> 32 GB and 1.8 TB/s: the fastest single consumer card for decode.

Canonical: https://thelocalbench.com/hardware/rtx-5090-32g · Data as of 2026-10-12 · The Local Bench (thelocalbench.com), data as of 2026-10-12, CC BY 4.0

Flags: [verified] checked against primary source · [sourced] cited · [estimate] formula · [sample] placeholder

## Specifications

| Field | Value | Flag | Source / note |
|---|---|---|---|
| memory_gb | 32 | sourced | https://www.nvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090/ |
| bandwidth_gbs | 1792 | sourced | https://en.wikipedia.org/wiki/GeForce_RTX_50_series |
| fp16_tflops | 209 | sample | Sample FP16 dense compute; pending vendor-sheet citation. |
| load_watts | 650 | sample | Sample wall power under load. |
| price_usd | 7199.99 | sourced | https://www.newegg.com/p/pl?d=rtx+5090+32gb+graphics+card&N=4131 |

## Models that run (default quant, 8k)

| Model | Quant | Memory GB | Verdict | Decode tok/s |
|---|---|---|---|---|
| [Llama 3.1 8B Instruct](/models/llama-3.1-8b.md) | Q4_K_M | 6.6 | fit | 179.4 [estimate] |
| [Qwen3 8B](/models/qwen3-8b.md) | Q4_K_M | 6.8 | fit | 172.4 [estimate] |
| [Qwen3 14B](/models/qwen3-14b.md) | Q4_K_M | 10.9 | fit | 104.0 [estimate] |
| [gpt-oss-20b](/models/gpt-oss-20b.md) | MXFP4 | 12.9 | fit | 469.6 [estimate] |
| [Mistral Small 3.2 24B](/models/mistral-small-3.2-24b.md) | Q4_K_M | 16.2 | fit | 68.6 [estimate] |
| [Gemma 3 27B IT](/models/gemma-3-27b.md) | Q4_K_M | 18.2 | fit | 60.9 [estimate] |
| [Qwen3 30B-A3B](/models/qwen3-30b-a3b.md) | Q4_K_M | 19.9 | fit | 382.4 [estimate] |
| [Qwen3 32B](/models/qwen3-32b.md) | Q4_K_M | 22.5 | fit | 49.1 [estimate] |
| [Llama 3.3 70B Instruct](/models/llama-3.3-70b.md) | Q4_K_M | 45.8 | offload | 2.1 [estimate] |
| [gpt-oss-120b](/models/gpt-oss-120b.md) | MXFP4 | 64.3 | offload | 21.6 [estimate] |