Qwen 3.5 35B
A large language model in Alibaba's Tongyi Qianwen (Qwen) series with a parameter scale of 35B. It requires 32GB of memory to run locally.
Timeline (last 90 days)
Perplexity currently offers three local orchestration models: Gemma 4 E4B (4-bit quantized, 6.6GB, requires 16GB of memory), Qwen 3.5 35B (requires 32GB of memory) and Perplexity's in-house 19GB model (requires 32GB of memory).
On the same machine and model, the decoding speed difference between low power mode and high power mode is 1.4 to 3.8 times; for example, running Qwen 3.5 35B on Ollama gives 51.9 tokens/s in low power and 107.1 tokens/s in high power.
Based on public data for high power plus MLX or Ollama, the decoding speed of different models on M4 Max is roughly a ladder as follows: Nemotron 3 Nano about 160 tokens/s, Qwen 3.5 35B at 107 to 130 tokens/s, Gemma 4 26B about 75 tokens/s, DeepSeek V4 Flash about 27 tokens/s, and dense 27B about 22 tokens/s.
All Facts (3)
Perplexity currently offers three local orchestration models: Gemma 4 E4B (4-bit quantized, 6.6GB, requires 16GB of memory), Qwen 3.5 35B (requires 32GB of memory) and Perplexity's in-house 19GB model (requires 32GB of memory).
50%UnverifiedOn the same machine and model, the decoding speed difference between low power mode and high power mode is 1.4 to 3.8 times; for example, running Qwen 3.5 35B on Ollama gives 51.9 tokens/s in low power and 107.1 tokens/s in high power.
50%UnverifiedBased on public data for high power plus MLX or Ollama, the decoding speed of different models on M4 Max is roughly a ladder as follows: Nemotron 3 Nano about 160 tokens/s, Qwen 3.5 35B at 107 to 130 tokens/s, Gemma 4 26B about 75 tokens/s, DeepSeek V4 Flash about 27 tokens/s, and dense 27B about 22 tokens/s.
50%