Compare models on your machine

Not generic leaderboards: every model is rated for your hardware, your tool and your workload — so you can see which one will actually feel better to use.

Hardware

Use case

Tool

Other apps

Context

Qwen3 8BRecursive Flint (40B-A2.5B)Qwen3.6 35B-A3BQwen3.5 35B-A3B

MacBook Pro M4 Pro 48 GB · Agentic coding / medium repository

Sort by

Recursive Flint (40B-A2.5B)MXFP4 · LM StudioComfortableSpeed 65–90 tok/sUses 22 GBFree 15 GB
Qwen3.6 35B-A3B5-bit · LM StudioComfortableSpeed 50–70 tok/sUses 25 GBFree 12 GB
Qwen3.5 35B-A3B5-bit · LM StudioComfortableSpeed 50–70 tok/sUses 25 GBFree 12 GB
Qwen3 8B6-bit · LM StudioAcceptableSpeed 17–25 tok/sUses 12 GBFree 25 GB

Model									Confidence
Recursive Flint (40B-A2.5B)MXFP4 · LM Studio	Comfortable	22 GB	15 GB	Excellent65–90 tok/s	Comfortable	128K	Strong	Strong	Medium confidence
Qwen3.6 35B-A3B5-bit · LM Studio	Comfortable	25 GB	12 GB	Excellent50–70 tok/s	Comfortable	128K	Strong	Strong	Medium confidence
Qwen3.5 35B-A3B5-bit · LM Studio	Comfortable	25 GB	12 GB	Excellent50–70 tok/s	Comfortable	128K	Good	Good	Medium confidence
Qwen3 8B6-bit · LM Studio	Acceptable	12 GB	25 GB	Adequate17–25 tok/s	Slow	32K	Fair	Fair	Medium confidence

Qwen3 8B

6-bit · LM Studio

Acceptable

Usable for agentic coding, but expect compromises: ingesting a prompt of 18K tokens takes ≈65 s and each step ≈40 s; across ~25 calls a task takes ≈18 min; generation (17–25 tok/s) is adequate but not ideal for agentic coding.

MemoryExcellent

≈25 GB headroom

GenerationFair

17–25 tok/s

Prompt processingPoor

≈65 s first prompt

ContextPoor

32K context

Tool compatibilityExcellent

Official

Workload suitabilityPoor

Poorly suited

Recursive Flint (40B-A2.5B)

MXFP4 · LM Studio

Comfortable

Yes — this should be comfortable for agentic coding.

MemoryExcellent

≈15 GB headroom

GenerationExcellent

65–90 tok/s

Prompt processingGood

≈20 s first prompt

ContextExcellent

32K context

Tool compatibilityExcellent

Official

Workload suitabilityGood

Suited

Qwen3.6 35B-A3B

5-bit · LM Studio

Comfortable

Yes — this should be comfortable for agentic coding.

MemoryGood

≈12 GB headroom

GenerationExcellent

50–70 tok/s

Prompt processingGood

≈20 s first prompt

ContextExcellent

32K context

Tool compatibilityExcellent

Official

Workload suitabilityGood

Suited

Qwen3.5 35B-A3B

5-bit · LM Studio

Comfortable

Yes — this should be comfortable for agentic coding.

MemoryGood

≈12 GB headroom

GenerationExcellent

50–70 tok/s

Prompt processingGood

≈20 s first prompt

ContextExcellent

32K context

Tool compatibilityExcellent

Official

Workload suitabilityGood

Suited