MLX Model Test Report

Generated 2026-02-21 15:18 · AFM MLX Backend · Apple M3 Ultra (512GB) · 3000 max tokens · temp 0.7

Models Tested
27
Passed
26
Failed
1
Best tok/s
337.8
Fastest Model
mlx-community/Qwen3-0.6B-4bit

Performance Ranking (by tokens/sec)

Click a row to jump to its full response below.

# Model Status Load (s) Tokens Gen Time (s) Tokens/sec Response Preview
1 mlx-community/Qwen3-0.6B-4bit OK 2 2878 8.52
337.8
 To explore calculus concepts from limits through multivariable calculus, let's break down the key ideas and connect
2 mlx-community/Llama-3.2-1B-Instruct-4bit OK 2 731 2.7
270.4
 **Limits** ================ Limits are a fundamental concept in multivariable calculus, which generalize the concept of
3 mlx-community/Qwen2.5-0.5B-Instruct-4bit OK 1 3114 13.91
223.8
 Calculus is the study of quantities or changes in quantities that can be described in terms of derivatives and integrals
4 mlx-community/LFM2-VL-3B-4bit OK 1 597 2.69
222.2
 Calculus is a branch of mathematics that deals with the study of continuous change. It comprises two major parts: differ
5 mlx-community/lille-130m-instruct-8bit OK 1 445 2.05
216.8
 Calculus is a branch of mathematics that deals with the study of limits in a strict mathematical notation. It involves d
6 mlx-community/LFM2-2.6B-4bit OK 1 1741 9.34
186.3
 Certainly! Let's explore the progression of calculus concepts from **limits** to **multivariable calculus**, buildi
7 mlx-community/exaone-4.0-1.2b-4bit OK 1 831 4.97
167.1
 Certainly! Below is a rigorous explanation of key concepts in calculus, starting from limits and progressing to multivar
8 mlx-community/Ling-mini-2.0-4bit OK 3 2354 15.18
155.1
 Certainly! Let's walk through **calculus concepts from limits through multivariable calculus**, using **rigorous ma
9 mlx-community/granite-4.0-h-tiny-4bit OK 2 982 8.3
118.3
 Certainly! Calculus is a fundamental branch of mathematics that deals with the study of change and motion. It provides a
10 mlx-community/SmolLM3-3B-4bit OK 2 1421 13.23
107.4
 Certainly, let's delve into the concept of limits from single-variable calculus and then transition into multivaria
11 mlx-community/gpt-oss-20b-MXFP4-Q4 OK 4 2209 23.51
94.0
 <|channel|>analysis<|message|>We need to explain calculus concepts from limits through multivariable calculu
12 mlx-community/gemma-3-4b-it-8bit OK 3 1333 15.53
85.8
 Okay, let's embark on a journey through calculus, building from the fundamental concept of limits to the complexiti
13 mlx-community/Qwen3-VL-4B-Instruct-4bit OK 2 2180 25.5
85.5
 Certainly. Below is a rigorous, mathematically precise exposition of calculus concepts, progressing from **limits** thro
14 hub/models--mlx-community--Qwen3-VL-4B-Instruct-4bit OK 2 2829 34.38
82.3
 Certainly. Below is a rigorous, mathematically precise exposition of calculus concepts, progressing from limits through
15 mlx-community/Qwen3-30B-A3B-4bit OK 6 3030 37.31
81.2
 ### **1. Limits and Continuity** #### **Definition of a Limit (Single Variable)** Let $ f: \mathbb{R} \to \mathbb{R} $,
16 mlx-community/gpt-oss-20b-MXFP4-Q8 OK 3 3284 40.89
80.3
 <|channel|>analysis<|message|>The user wants an explanation of calculus concepts from limits through multiva
17 mlx-community/Qwen3-VL-4B-Instruct-8bit OK 2 2209 34.35
64.3
 Certainly. Below is a rigorous, step-by-step exposition of calculus concepts, progressing from the foundational notion o
18 mlx-community/NVIDIA-Nemotron-3-Nano-30B-A3B-4bit OK 6 2703 51.5
52.5
 Below is a compact, yet fully rigorous, roadmap of calculus from the **ε–δ definition of a limit** through the core idea
19 mlx-community/GLM-4.7-Flash-4bit OK 7 2608 51.11
51.0
 Here is an explanation of calculus concepts, organized from single-variable calculus (Limits, Derivatives, Integrals) to
20 mlx-community/JoyAI-LLM-Flash-4bit-DWQ OK 9 721 16.86
42.8
 ## **Calculus: From Limits to Multivariable Calculus** ### **1. Limits & Continuity** $$ \lim_{x \to a} f(x) = L \i
21 mlx-community/Qwen3-Coder-Next-4bit OK 16 3088 76.06
40.6
 # A Rigorous Introduction to Calculus: From Limits to Multivariable Calculus ## 1. Limits ### 1.1 ε-δ Definition of a
22 mlx-community/Apertus-8B-Instruct-2509-4bit OK 2 50 1.36
36.8
 n uhua spy " word of,opollu ब所以uck в مناхаيااس.स�كي هاف ای 대ر나무 goychẹpర (se nonhanak the movie stark thaklakah kyy
23 mlx-community/MiniMax-M2.5-5bit OK 47 1991 63.03
31.6
 # Calculus: From Limits to Multivariable Calculus ## 1. Limits and Continuity ### 1.1 Limit of a Function For a functi
24 mlx-community/MiniMax-M2.5-6bit OK 59 2832 98.39
28.8
 # Calculus: From Limits to Multivariable Calculus ## 1. Limits and Continuity ### 1.1 Limit of a Function **Definitio
25 mlx-community/GLM-5-4bit OK 129 3394 341.13
9.9
 Here is a rigorous exposition of calculus concepts, ranging from the foundational definition of a limit to the generaliz
26 mlx-community/Qwen3.5-397B-A17B-4bit OK 67 3227 671.02
4.8
 The following is a structured exposition of calculus, progressing from the foundational definitions of single-variable a

Failed Models

ModelErrorLoad Time (s)
mlx-community/gemma-3n-E2B-it-lm-4bit mlx-community/gemma-3n-E2B-it-lm-4bit: Unsupported model type: gemma3n 1

Full Model Responses

▶ mlx-community/Qwen3-0.6B-4bit 2878 tokens · 337.8 tok/s

▶ mlx-community/Llama-3.2-1B-Instruct-4bit 731 tokens · 270.4 tok/s

▶ mlx-community/Qwen2.5-0.5B-Instruct-4bit 3114 tokens · 223.8 tok/s

▶ mlx-community/LFM2-VL-3B-4bit 597 tokens · 222.2 tok/s

▶ mlx-community/lille-130m-instruct-8bit 445 tokens · 216.8 tok/s

▶ mlx-community/LFM2-2.6B-4bit 1741 tokens · 186.3 tok/s

▶ mlx-community/exaone-4.0-1.2b-4bit 831 tokens · 167.1 tok/s

▶ mlx-community/Ling-mini-2.0-4bit 2354 tokens · 155.1 tok/s

▶ mlx-community/granite-4.0-h-tiny-4bit 982 tokens · 118.3 tok/s

▶ mlx-community/SmolLM3-3B-4bit 1421 tokens · 107.4 tok/s

▶ mlx-community/gpt-oss-20b-MXFP4-Q4 2209 tokens · 94.0 tok/s

▶ mlx-community/gemma-3-4b-it-8bit 1333 tokens · 85.8 tok/s

▶ mlx-community/Qwen3-VL-4B-Instruct-4bit 2180 tokens · 85.5 tok/s

▶ hub/models--mlx-community--Qwen3-VL-4B-Instruct-4bit 2829 tokens · 82.3 tok/s

▶ mlx-community/Qwen3-30B-A3B-4bit 3030 tokens · 81.2 tok/s

▶ mlx-community/gpt-oss-20b-MXFP4-Q8 3284 tokens · 80.3 tok/s

▶ mlx-community/Qwen3-VL-4B-Instruct-8bit 2209 tokens · 64.3 tok/s

▶ mlx-community/NVIDIA-Nemotron-3-Nano-30B-A3B-4bit 2703 tokens · 52.5 tok/s

▶ mlx-community/GLM-4.7-Flash-4bit 2608 tokens · 51.0 tok/s

▶ mlx-community/JoyAI-LLM-Flash-4bit-DWQ 721 tokens · 42.8 tok/s

▶ mlx-community/Qwen3-Coder-Next-4bit 3088 tokens · 40.6 tok/s

▶ mlx-community/Apertus-8B-Instruct-2509-4bit 50 tokens · 36.8 tok/s

▶ mlx-community/MiniMax-M2.5-5bit 1991 tokens · 31.6 tok/s

▶ mlx-community/MiniMax-M2.5-6bit 2832 tokens · 28.8 tok/s

▶ mlx-community/GLM-5-4bit 3394 tokens · 9.9 tok/s

▶ mlx-community/Qwen3.5-397B-A17B-4bit 3227 tokens · 4.8 tok/s