Which local LLMs can the GeForce RTX 3090 run?
Assumes 32 GB of DDR5-5600 system RAM, Windows with this GPU driving the display, 8k context and an f16 KV cache.
Specs
- VRAM
- 24 GB
- Memory bandwidth
- 936.0 GB/s
- FP16 compute
- 35.6 TFLOPS
- Launch year
- 2020
- Price
- $1,050 street (as of 2026-08-09)
Verdicts at a glance
At Q4_K_M (or the closest available quant) with 8k context.
- Runs great23 modelsRuns well1 modelRuns slowly0 modelsWon't run5 models2 more models run at a lower quant.
How to read the verdicts
- Runs great
- Fully on the GPU at 20 tok/s or more
- Runs well
- Fully on the GPU at 8–20 tok/s; MoE experts in system RAM at 20 tok/s or more; or at least 90% on the GPU at 8 tok/s or more
- Runs slowly
- 2–8 tok/s; CPU-only; less than 90% on the GPU; or MoE experts in system RAM below 20 tok/s
- Won't run
- Does not fit, or under 2 tok/s
Every model on the GeForce RTX 3090
Dense models split between GPU and CPU slow down sharply — the CPU side sets the pace. MoE models that keep only their experts in system RAM degrade far more gently.
Scroll sideways to see every column.
| Model | Verdict | Speed | Quant | Memory | Runs as | Context | Notes |
|---|---|---|---|---|---|---|---|
| EXAONE 4.0 1.2B | Runs great | est. 498.9 tok/s439.0–558.8calibrated estimate ±12% | Q4_K_M | 2.5 / 24.0 GB | Full GPU | up to 64k | — |
| HyperCLOVA X SEED 1.5B | Runs great | est. 360.9 tok/s317.6–404.2calibrated estimate ±12% | Q4_K_M | 3.0 / 24.0 GB | Full GPU | up to 16k | — |
| Qwen3.5 35B-A3B | Runs great | est. 204.8 tok/s163.9–245.8calibrated estimate ±20% | Q4_K_M | 24.0 / 24.0 GB | Full GPU | up to 8k | — |
| gpt-oss-20b | Runs greatDetails | 162.0 tok/s8k estimate 184.2 tok/s (147.3–221.0, calibrated estimate ±20%)measured (1 run, 2k context) | MXFP4 | 13.5 / 24.0 GB | Full GPU | up to 128k |
|
| Qwen3 30B-A3B (2507) | Runs greatDetails | est. 149.7 tok/s119.8–179.7calibrated estimate ±20% | Q4_K_M | 20.5 / 24.0 GB | Full GPU | up to 32k | — |
| Gemma 4 26B-A4B | Runs great | est. 145.7 tok/s116.6–174.9calibrated estimate ±20% | Q4_K_M | 18.5 / 24.0 GB | Full GPU | up to 64k | — |
| Kanana 1.5 15.7B-A3B | Runs great | est. 137.3 tok/s109.8–164.7calibrated estimate ±20% | Q4_K_M | 12.7 / 24.0 GB | Full GPU | up to 32k | — |
| DeepSeek R1 Distill Llama 8B | Runs great | est. 109.3 tok/s96.2–122.4calibrated estimate ±12% | Q4_K_M | 7.2 / 24.0 GB | Full GPU | up to 64k |
|
| Kanana 1.5 8B | Runs great | est. 109.3 tok/s96.2–122.4calibrated estimate ±12% | Q4_K_M | 7.2 / 24.0 GB | Full GPU | up to 32k | — |
| Llama 3.1 8B | Runs greatDetails | est. 109.3 tok/s96.2–122.4calibrated estimate ±12% | Q4_K_M | 7.2 / 24.0 GB | Full GPU | up to 64k | — |
| Qwen3.5 9B | Runs great | est. 106.7 tok/s93.9–119.5calibrated estimate ±12% | Q4_K_M | 7.3 / 24.0 GB | Full GPU | up to 256k | — |
| Qwen3 8B | Runs great | est. 105.0 tok/s92.4–117.6calibrated estimate ±12% | Q4_K_M | 7.4 / 24.0 GB | Full GPU | up to 32k | — |
| Gemma 4 12B | Runs great | est. 85.6 tok/s75.3–95.8calibrated estimate ±12% | Q4_K_M | 8.8 / 24.0 GB | Full GPU | up to 128k | — |
| Gemma 3 12B | Runs greatDetails | est. 83.6 tok/s73.6–93.6calibrated estimate ±12% | Q4_K_M | 9.0 / 24.0 GB | Full GPU | up to 128k | — |
| HyperCLOVA X SEED Think 14B | Runs great | est. 64.3 tok/s45.0–83.5theoretical estimate ±30% | Q4_K_M | 11.4 / 24.0 GB | Full GPU | up to 64k |
|
| Qwen3 14B | Runs greatDetails | est. 63.4 tok/s55.7–71.0calibrated estimate ±12% | Q4_K_M | 11.5 / 24.0 GB | Full GPU | up to 32k | — |
| Phi-4 | Runs great | est. 61.1 tok/s53.7–68.4calibrated estimate ±12% | Q4_K_M | 11.9 / 24.0 GB | Full GPU | up to 16k |
|
| Mistral Small 3.2 24B | Runs great | est. 41.8 tok/s36.8–46.8calibrated estimate ±12% | Q4_K_M | 16.8 / 24.0 GB | Full GPU | up to 32k | — |
| Gemma 3 27B | Runs great | est. 38.0 tok/s33.5–42.6calibrated estimate ±12% | Q4_K_M | 18.4 / 24.0 GB | Full GPU | up to 64k | — |
| Qwen3.5 27B | Runs great | est. 37.1 tok/s32.7–41.6calibrated estimate ±12% | Q4_K_M | 18.8 / 24.0 GB | Full GPU | up to 64k | — |
| EXAONE 4.0 32B | Runs great | est. 33.0 tok/s29.0–36.9calibrated estimate ±12% | Q4_K_M | 21.1 / 24.0 GB | Full GPU | up to 32k |
|
| EXAONE 4.5 33B | Runs great | est. 31.8 tok/s28.0–35.6calibrated estimate ±12% | Q4_K_M | 21.8 / 24.0 GB | Full GPU | up to 16k |
|
| Qwen3 32B | Runs great | est. 29.9 tok/s26.3–33.5calibrated estimate ±12% | Q4_K_M | 23.1 / 24.0 GB | Full GPU | up to 8k | — |
| DeepSeek R1 Distill Qwen 32B | Runs well | est. 29.8 tok/s26.2–33.4calibrated estimate ±12% | Q4_K_M | 23.2 / 24.0 GB | Full GPU | up to 8k |
|
| Qwen3.5 122B-A10B | Won't runTry UD-Q2_K_XL (heavy quality loss): Runs slowly | est. 20.4 tok/s14.3–26.5theoretical estimate ±30% | UD-Q2_K_XL | 24.0 / 24.0 GB + 20.2 GB RAM | Partial offload | up to 128k |
|
| Llama 3.3 70B | Won't runTry IQ4_XS: Runs slowly | est. 2.2 tok/s1.7–2.6calibrated estimate ±20% | IQ4_XS | 24.0 / 24.0 GB + 17.8 GB RAM | Partial offload | up to 8k | — |
| gpt-oss-120b | Won't runDetails | — | MXFP4 | needs 64.3 GB | — | — |
|
| Solar Open 100B | Won't run | — | Q4_K_M | needs 64.4 GB | — | — |
|
| Solar Open 2 250B | Won't run | — | IQ4_XS | needs 137.2 GB | — | — |
|
Reasoning models spend extra tokens thinking, so their speed thresholds are 1.5× stricter (30 / 12 / 3 tok/s).
Measured results on the GeForce RTX 3090
Public benchmarks we calibrate against. Their conditions (context, backend, flags) can differ from the estimates above.
| Model | Quant | Backend | Context | Prompt (tok/s) | Generation (tok/s) | Flags | Source | Measured |
|---|---|---|---|---|---|---|---|---|
| llama-2-7b | Q4_0 | llama.cpp | 512 | 5,175.0 | 158.2 | — | github.com | 2025-08-01 |
| gpt-oss-20b | MXFP4 | llama.cpp | 2k | 5,143.0 | 162.0 | — | github.com | 2025-08-15 |
| Qwen3 8B | Q4_K_XL | llama.cpp | 16k | — | 87.5 | — | hardware-corner.net | 2026-08-09 |
| Qwen3 30B-A3B (2507) | Q4_K_XL | llama.cpp | 16k | — | 113.8 | — | hardware-corner.net | 2026-08-09 |
Frequently asked questions
What is the largest model that runs entirely on the GeForce RTX 3090?
Qwen3.5 35B-A3B at Q4_K_M (a 22.63 GB file) fits entirely in 24 GB with 8k context, at est. 204.8 tok/s (163.9–245.8, calibrated estimate ±20%).
How many local LLMs run well on the GeForce RTX 3090?
At Q4_K_M with 8k context, out of 29 tracked models: 23 run great, 1 runs well, 0 run slowly and 5 won't run.
Can the GeForce RTX 3090 run a 70B model like Llama 3.3 70B?
Runs slowly — Partial offload, IQ4_XS: est. 2.2 tok/s (1.7–2.6, calibrated estimate ±20%).
Speeds are estimates from memory bandwidth, calibrated against public benchmarks, and each one comes with an error band and a confidence label. Real results vary with drivers, backend, context length and thermals.