What can the Snapdragon X Elite (Adreno X1-85) run?
With 16 GB, 3 of 12 popular open models fit. The best pick is Qwen3 4B at about 30 tok/s, about reading speed.
ollama run qwen3:4bFirst time? Install Ollama, then paste that into a terminal. Is this your machine? Check it on the homepage.
Every model the Snapdragon X Elite (Adreno X1-85) 16 GB can run
3 of 12 fit. Best pick: Qwen3 4B.
- Qwen3 4BQuick everyday chat and summaries~30 tok/sUsable
qwen3:4b - Qwen3 8BA solid all-rounder for chat and writing~14 tok/sUsable
qwen3:8b - Gemma 3 12BReads images as well as text~9 tok/sSlow
gemma3:12b - Qwen3 14BNoticeably smarter at reasoning and codeneeds 11 GBToo big
qwen3:14b - gpt-oss 20BOpenAI's open model. Fast, strong reasoningneeds 16 GBToo big
gpt-oss:20b - Gemma 3 27BGoogle's best open model for its sizeneeds 19 GBToo big
gemma3:27b - Qwen3 30B-A3B30B smarts at small-model speedneeds 21 GBToo big
qwen3:30b - Qwen3 32BStrong coding and long answersneeds 22 GBToo big
qwen3:32b - Llama 3.3 70BClose to the big cloud modelsneeds 45 GBToo big
llama3.3:70b - gpt-oss 120BOpenAI's largest open modelneeds 67 GBToo big
gpt-oss:120b - Qwen3 235B-A22BAlibaba's flagship open model, for multi-GPU rigsneeds 144 GBToo big
qwen3:235b - DeepSeek R1 671BThe full DeepSeek R1, for the biggest rigsneeds 406 GBToo big
deepseek-r1:671b
Every model the Snapdragon X Elite (Adreno X1-85) 32 GB can run
6 of 12 fit. Best pick: gpt-oss 20B.
- Qwen3 4BQuick everyday chat and summaries~30 tok/sUsable
qwen3:4b - Qwen3 8BA solid all-rounder for chat and writing~14 tok/sUsable
qwen3:8b - Gemma 3 12BReads images as well as text~9 tok/sSlow
gemma3:12b - Qwen3 14BNoticeably smarter at reasoning and code~8 tok/sSlow
qwen3:14b - gpt-oss 20BOpenAI's open model. Fast, strong reasoning~24 tok/sUsable
gpt-oss:20b - Gemma 3 27BGoogle's best open model for its size~4 tok/sSlow
gemma3:27b - Qwen3 30B-A3B30B smarts at small-model speedneeds 21 GBToo big
qwen3:30b - Qwen3 32BStrong coding and long answersneeds 22 GBToo big
qwen3:32b - Llama 3.3 70BClose to the big cloud modelsneeds 45 GBToo big
llama3.3:70b - gpt-oss 120BOpenAI's largest open modelneeds 67 GBToo big
gpt-oss:120b - Qwen3 235B-A22BAlibaba's flagship open model, for multi-GPU rigsneeds 144 GBToo big
qwen3:235b - DeepSeek R1 671BThe full DeepSeek R1, for the biggest rigsneeds 406 GBToo big
deepseek-r1:671b
Can the Snapdragon X Elite (Adreno X1-85) run it?
Can the Snapdragon X Elite (Adreno X1-85) run gpt-oss 20B?
Yes, with 32 GB or more. gpt-oss 20B needs about 16 GB and the Snapdragon X Elite (Adreno X1-85) with 32 GB of system memory leaves about 19 GB for models. Expect about 24 tok/s, about reading speed. The 16 GB version doesn't have room.
Can the Snapdragon X Elite (Adreno X1-85) run Gemma 3 27B?
Yes, with 32 GB or more. Gemma 3 27B needs about 19 GB and the Snapdragon X Elite (Adreno X1-85) with 32 GB of system memory leaves about 19 GB for models. Expect about 4 tok/s, slow, but it works. The 16 GB version doesn't have room.
Can the Snapdragon X Elite (Adreno X1-85) run Qwen3 30B-A3B?
No. Qwen3 30B-A3B needs about 21 GB and the Snapdragon X Elite (Adreno X1-85) has at most 32 GB of system memory, of which about 19 GB is usable for models. The most capable model it runs well is gpt-oss 20B, at about 24 tok/s.
Real runs on the Snapdragon X Elite (Adreno X1-85)
Loading measured runs…
Generating a token reads the model's active weights from memory once, so speed is roughly memory bandwidth (135 GB/s) divided by the active weight size, with efficiency calibrated on our own rack. Sizes are Ollama's default 4-bit downloads plus about 1.2 GB for context. Built-in graphics share system memory, so we count about 60% of it. The easiest start is a small model in your browser in the Studio. Real results are usually within about a third. Measured runs are on the leaderboard.
The Used desktop + RTX 3060 12 GB runs Qwen3 14B at about 21 tok/s, from $600 used.