Local AI Advisor

Can MacBook Pro M3 Max 128 GB run Recursive Flint (40B-A2.5B)?

128 GB unified · 400 GB/s memory bandwidth · Recursive Flint (40B-A2.5B) has 39.8B parameters (2.5B active per token).

Short answer

Yes, excellent for chat and comfortable or better for every workload we test.

Best quantization for chat: MXFP4 (20 GB of weights) on LM Studio. Generation: 130–170 tok/s, leaving about 97 GB free for your other apps.

How it rates for each workload

Same model and machine, different jobs. Agents and long documents send far bigger prompts than chat, so they need much more speed and memory.

WorkloadVerdictQuantGenerationFirst replyContext
Chatwith Open WebUIExcellentMXFP4130–170 tok/sunder 1 s8KDetails
Coding questionswith ContinueExcellentMXFP4120–160 tok/s≈3 s16KDetails
Coding in a repositorywith AiderComfortableMXFP4100–140 tok/s≈8 s32KDetails
Agentic codingwith OpenCodeComfortableMXFP495–130 tok/s≈10 s32KDetails
Long documents (64K)with Open WebUIExcellentMXFP465–85 tok/s≈30 s64KDetails

Speeds are estimates calibrated against measured benchmarks on similar hardware. “First reply” is the time to read the workload’s opening prompt. How we calculate

Different tool, context size or workload?

These pages use typical settings. The full check models your exact setup.