Local model performance on your hardware

Find which open-weight LLMs fit in your GPU or Mac and compare the speeds people report on setups like yours.

What runs on your hardware?

Pick your GPU or Mac, then read the speeds people reported on it, or estimate which models fit and how fast they run.

Free to use. No account needed. Memory estimates and community measurements are labelled separately.

Looking for a particular model?

Search for a model to see its reported speeds across GPUs and Macs.

Local model performance reports from the community

These are individual setups, not a controlled benchmark. Compare GPU count, quantization, context and offloading before comparing speeds. How to read a report →

GPU: NVIDIA RTX 3070

Compare setup details (1 active)

QuantizationEngineRecorded GPU countReported context (tokens)

Exact recorded values. Context may be a configured limit; matching filters does not establish identical prompts, offloading or concurrency.

Qwen3.6 35B (3B active)

NVIDIA RTX 3070 · llama.cpp · 128,000 ctx

generation:: 30.0 tokens/s
quant:: Q5_K_S (GGUF)

Reported by the source; GPU count, offloading and concurrent requests can change these figures. Prompt-processing speed is shown only when the source reports it; generation speed cannot tell us time to first token. Check the full setup before comparing.

User reports Qwen3.6-35B-A3B at about 30 t/s with Q5_K_S at 128k context on an RTX 3070 8GB with 64GB DDR4. Setup is llama.cpp. Q4_K_XL reached 32 t/s, and IQ4_XS gave 25-30 t/s at 32k context. The user found that larger quants (Q4_K_XL, Q5_K_S) gave better speed than the smaller Q4 (IQ4_XS).

Showing 1–1 of 1

Page 1 of 1

Community benchmarks snapshot

Records by GPU

AMD Strix Halo 128GB295NVIDIA RTX 3090190NVIDIA RTX 5090134NVIDIA DGX Spark78M5 Max 48GB65NVIDIA RTX Pro 6000 Blackwell62NVIDIA RTX 5060 Ti 16GB58NVIDIA RTX 3060 12GB54AMD Radeon AI PRO R9700 32GB50M4 Max 128GB49

Records by model

Qwen3.81022

Qwen3.6238

Gemma 4150

DeepSeek V4 Flash147

Qwen3.598

Qwen353

other514

Records by engine

llama.cpp881

vLLM189

MLX101

Strata70

Ollama62

other352

Use cases

coding 558agentic 358long-context 272tool-use 153vision 109summarization 53math 43creative-writing 37multilingual 23text-generation 9rp 8reasoning 3

coding558agentic358long-context272tool-use153vision109summarization53math43creative-writing37

Median generation t/s by GPU

On Qwen3.8 27B at 4-bit, plain single-GPU runs. Full ranking

RTX 509082RTX Pro 600067M5 Ultra 256GB50RX 7900 XTX48RTX 309036RTX 5090 Laptop 24GB34RTX 409032V100 32GB31Radeon AI PRO R9700 32GB30RX 7800 XT 16GB30

Prompt processing, same model

M5 Ultra 256GB1800RX 7900 XTX530RTX 3090561RTX 40901949V100 32GB941RTX 3080 20GB997Arc Pro B70512M1 Ultra 128GB153Arc Pro B60 24GB380M1 Max 32GB82

Reports by model size

Qwen3.8 27B566Qwen3.8 125B · 6B active372DeepSeek V4 Flash 284B · 13B active130Qwen3.6 35B · 3B active129Qwen3.6 27B72Gemma 4 26B · 4B active36GLM-5.3 320B · 18B active33DeepSeek V4.1 Flash 552B · 16B active28

Quants

Q4_K_M296UD-Q4_K_XL141NVFP4120Q8_0984-bit81Q4_K_XL71IQ4_XS70MXFP458IQ3_XXS52Q450