MLX Model Test Report

Generated 2026-02-18 00:49 · AFM MLX Backend · Apple M4 Ultra (512GB) · 3000 max tokens · temp 0.7

Models Tested
30
Passed
25
Failed
5
Best tok/s
453.8
Fastest Model
mlx-community/lille-130m-instruct-8bit

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/lille-130m-instruct-8bit OK 1 3973 8.76
453.8
 Calculus is a fundamental concept in calculus that deals with the study of limits, which are sets of real numbers, and t
2 mlx-community/Qwen2.5-0.5B-Instruct-4bit OK 2 3820 8.7
438.9
 Calculus is a branch of mathematics that deals with the study of rates of change and the accumulation of quantities. It
3 mlx-community/Qwen3-0.6B-4bit OK 2 1769 5.14
344.1
 <think> Okay, let's see. The user wants me to explain calculus concepts from limits through multivariable cal
4 mlx-community/Llama-3.2-1B-Instruct-4bit OK 3 871 2.74
318.4
 **Limits** Limits are a fundamental concept in multivariable calculus. They allow us to define the behavior of function
5 mlx-community/LFM2-VL-3B-4bit OK 1 651 3.19
204.3
 Calculus, at its core, is the study of change. It is divided into two main branches: differential calculus and integral
6 mlx-community/LFM2-2.6B-4bit OK 1 2056 11.14
184.5
 Certainly. Below is a rigorous mathematical explanation of calculus concepts, progressing from **limits** to **multivari
7 mlx-community/exaone-4.0-1.2b-4bit OK 1 685 3.89
175.9
 Certainly! Below is a rigorous explanation of key concepts in calculus, progressing from limits to multivariable calculu
8 mlx-community/Ling-mini-2.0-4bit OK 3 989 6.38
155.1
 Certainly! Let's explore the foundational concepts of calculus, starting from limits and progressing through to mul
9 mlx-community/SmolLM3-3B-4bit OK 2 1499 10.64
140.9
 <think> </think> Certainly! Let's start with the basics and build up through the concepts. Here's
10 mlx-community/granite-4.0-h-tiny-4bit OK 2 641 5.76
111.2
 Certainly! Let's start with the basics of calculus and gradually build up to multivariable calculus concepts, using
11 mlx-community/gpt-oss-20b-MXFP4-Q4 OK 4 1631 16.26
100.3
 <|channel|>analysis<|message|>We need to explain calculus concepts from limits to multivariable calculus wit
12 mlx-community/Qwen3-VL-4B-Instruct-4bit OK 2 1887 23.1
81.7
 Certainly. Below is a rigorous, mathematically precise exposition of calculus concepts, progressing from limits through
13 hub/models--mlx-community--Qwen3-VL-4B-Instruct-4bit OK 3 1885 23.2
81.3
 Certainly. Below is a rigorous, mathematically precise exposition of calculus concepts, progressing from limits through
14 mlx-community/gpt-oss-20b-MXFP4-Q8 OK 3 2011 25.86
77.8
 <|channel|>analysis<|message|>We need to explain calculus concepts from limits through multivariable calculu
15 mlx-community/Qwen3-30B-A3B-4bit OK 6 2272 29.79
76.3
 <think> Okay, I need to explain calculus concepts from limits through multivariable calculus with rigorous mathema
16 mlx-community/gemma-3n-E2B-it-lm-4bit OK 3 2063 28.64
72.0
 Okay, let's embark on a journey through calculus, starting with limits, and progressing through the core concepts o
17 mlx-community/Qwen3-VL-4B-Instruct-8bit OK 2 1907 31.88
59.8
 Certainly. Below is a rigorous, mathematically precise exposition of calculus concepts, progressing from limits through
18 mlx-community/GLM-4.7-Flash-4bit OK 6 2387 44.11
54.1
 The user wants an explanation of calculus concepts, ranging from limits to multivariable calculus, using rigorous mathem
19 mlx-community/NVIDIA-Nemotron-3-Nano-30B-A3B-4bit OK 6 2398 48.24
49.7
 The user wants an explanation of calculus concepts from limits through multivariable calculus with rigorous mathematical
20 mlx-community/Qwen3-Coder-Next-4bit OK 14 1858 44.56
41.7
 # A Rigorous Overview of Calculus: From Limits to Multivariable Calculus ## 1. Limits ### 1.1 ε-δ Definition of a Limi
21 mlx-community/JoyAI-LLM-Flash-4bit-DWQ OK 10 1584 44.2
35.8
 ## **1. Limits & Continuity** ### **1.1 Limit Definition** Let $f: D \to \mathbb{R}$ where $D \subseteq \mathbb{R}$
22 mlx-community/MiniMax-M2.5-5bit OK 49 1300 40.53
32.1
 We are going to explain calculus concepts from limits through multivariable calculus with rigorous mathematical notation
23 mlx-community/Apertus-8B-Instruct-2509-4bit OK 2 19 0.61
31.1
 l l ,las delantero o que - del deoothaveamente todoshheterap wordsasunaلكلك
24 mlx-community/MiniMax-M2.5-6bit OK 58 1887 66.46
28.4
 We are going to explain calculus concepts from limits through multivariable calculus with rigorous mathematical notation
25 mlx-community/GLM-5-4bit OK 125 2325 220.91
10.5
 The user wants an explanation of calculus concepts, starting from limits and going all the way through multivariable cal

Failed Models

ModelErrorLoad Time (s)
lmstudio-community/gemma-3n-E4B-it-MLX-4bit Model loading stalled (timeout) 300
mlx-community/gemma-3-4b-it-8bit 3
mlx-community/Mixtral-8x7B-Instruct-v0.1-4bit mlx-community/Mixtral-8x7B-Instruct-v0.1-4bit: Unsupported model type: mixtral 300
mlx-community/Qwen3.5-397B-A17B-4bit mlx-community/Qwen3.5-397B-A17B-4bit: Unsupported model type: qwen3_5_moe 300
sentence-transformers/all-MiniLM-L6-v2 sentence-transformers/all-MiniLM-L6-v2: Unsupported model type: bert 300

Full Model Responses

▶ mlx-community/lille-130m-instruct-8bit 3973 tokens · 453.8 tok/s

▶ mlx-community/Qwen2.5-0.5B-Instruct-4bit 3820 tokens · 438.9 tok/s

▶ mlx-community/Qwen3-0.6B-4bit 1769 tokens · 344.1 tok/s

▶ mlx-community/Llama-3.2-1B-Instruct-4bit 871 tokens · 318.4 tok/s

▶ mlx-community/LFM2-VL-3B-4bit 651 tokens · 204.3 tok/s

▶ mlx-community/LFM2-2.6B-4bit 2056 tokens · 184.5 tok/s

▶ mlx-community/exaone-4.0-1.2b-4bit 685 tokens · 175.9 tok/s

▶ mlx-community/Ling-mini-2.0-4bit 989 tokens · 155.1 tok/s

▶ mlx-community/SmolLM3-3B-4bit 1499 tokens · 140.9 tok/s

▶ mlx-community/granite-4.0-h-tiny-4bit 641 tokens · 111.2 tok/s

▶ mlx-community/gpt-oss-20b-MXFP4-Q4 1631 tokens · 100.3 tok/s

▶ mlx-community/Qwen3-VL-4B-Instruct-4bit 1887 tokens · 81.7 tok/s

▶ hub/models--mlx-community--Qwen3-VL-4B-Instruct-4bit 1885 tokens · 81.3 tok/s

▶ mlx-community/gpt-oss-20b-MXFP4-Q8 2011 tokens · 77.8 tok/s

▶ mlx-community/Qwen3-30B-A3B-4bit 2272 tokens · 76.3 tok/s

▶ mlx-community/gemma-3n-E2B-it-lm-4bit 2063 tokens · 72.0 tok/s

▶ mlx-community/Qwen3-VL-4B-Instruct-8bit 1907 tokens · 59.8 tok/s

▶ mlx-community/GLM-4.7-Flash-4bit 2387 tokens · 54.1 tok/s

▶ mlx-community/NVIDIA-Nemotron-3-Nano-30B-A3B-4bit 2398 tokens · 49.7 tok/s

▶ mlx-community/Qwen3-Coder-Next-4bit 1858 tokens · 41.7 tok/s

▶ mlx-community/JoyAI-LLM-Flash-4bit-DWQ 1584 tokens · 35.8 tok/s

▶ mlx-community/MiniMax-M2.5-5bit 1300 tokens · 32.1 tok/s

▶ mlx-community/Apertus-8B-Instruct-2509-4bit 19 tokens · 31.1 tok/s

▶ mlx-community/MiniMax-M2.5-6bit 1887 tokens · 28.4 tok/s

▶ mlx-community/GLM-5-4bit 2325 tokens · 10.5 tok/s