[whatmodel.sh](../../README.md) / Qwen3.5 2B
# Qwen3.5 2B
Alibaba · released 2026-03 · 2.27B params · apache-2.0
**score 47.2** ±0.6 · #426 of 732 · from 109 benchmark results across 13 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Qwen3.5 2B among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁●▆█▄▅▅▄▄▂▁ #290 of 313 9.2% 10
coding ▁▁▁●▅▇█▇▅▄▃▁ #449 of 469 8.5% 7
chat & writing ▁▁▁▃▆▄●█▆▇▇▃ #253 of 402 1.1% 1
math ▁▁▃▅●▆▄█▄▂▁▁ #308 of 431 26% 13
vision & documents ▁▁▂▂▁●▄▂▆▇█▄ #120 of 153 14% 20
web search ▅·▅●█·▃▃▃·▅▃ #10 of 15* 0.8% 2
agent tasks · #290 of 313 · 9.2% of its score
Beats [Bielik 11B v2.3 Instruct](bielik-11b-v2-3-instruct.md), [Mixtral 8x22B Instruct v0.1](mixtral-8x22b-instruct-v0-1.md), [GLM-4 9B Chat](glm-4-9b-chat.md) here, though they rank higher overall. Trails [Granite 4.0 H 350M](granite-4-0-h-350m.md), [Qwen3 0.6B](qwen3-0-6b.md), [Qwen2.5 1.5B Instruct](qwen2-5-1-5b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
BFCL V4 (December 2025 archive) 43.6% ▄·●·▄▄▄▄·▄·▄ beats 1/8 medium · avg 43% 1.9% at its level · several runs
SWE-bench Verified 5.0% ▄··▄▄●··▄·▄█ beats 3/8 very hard · avg 14% 1.7% sharp · small run · several runs
SWE-bench Pro (split unspecified) 0.8% █●···▂···▄·▂ beats 4/8 very hard · avg 8.0% 1.2% sharp · too hard for it · small run · s…
MCP-Atlas (public set) 14.8% ●·▁▂▂·▁▃█▆▆▂ beats 1/39 hard · avg 21% 1.0% typical
SWE-bench Multilingual 5.0% ●··▁▃▁▅██▂▂▁ beats 0/65 hard · avg 18% 0.9% typical
τ²-bench aggregate (maker) 48.8% ▁▁▁·▁●▃▃▃▄█▁ beats 5/39 medium · avg 46% 0.9% at its level · small run
BrowseComp-ZH 18.2% ▄▄·▄▄▄·●···█ beats 5/8 very hard · avg 13% 0.6% noisy
τ³-bench Banking 2.1% ▂●·▅█▂▃▃▃▆▂▅ beats 1/24 very hard · avg 3.6% 0.4% blunt · too hard for it
τ³-bench 4.1% ●·▃·█·▃··▃·▃ beats 0/7 very hard · avg 13% 0.4% typical
BrowseComp 3.1% ●···█·▃···▃▃ beats 0/7 very hard · avg 11% 0.3% noisy
coding · #449 of 469 · 8.5% of its score
Trails [Falcon3 1B Instruct](falcon3-1b.md), [Granite 3.0 1B-A400M Instruct](granite-3-0-1b-a400m.md), [Gemma 1.1 2B IT](gemma-1-1-2b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LiveCodeBench v6 (window unspecif… 20.2% ●····█·▃▅▃·▃ beats 0/8 medium · avg 38% 2.1% at its level
SWE-bench Verified 5.0% ▄··▄▄●··▄·▄█ beats 3/8 very hard · avg 14% 1.7% sharp · small run · several runs
LiveCodeBench Pro 2025Q2 (Easy) 10.3% ●···█▄··▄█▄▄ beats 0/8 hard · avg 33% 1.4% at its level
SWE-bench Pro (split unspecified) 0.8% █●···▂···▄·▂ beats 4/8 very hard · avg 8.0% 1.2% sharp · too hard for it · small run · s…
SWE-bench Multilingual 5.0% ●··▁▃▁▅██▂▂▁ beats 0/65 hard · avg 18% 0.9% typical
SciCode (level unspecified) 6.0% ●·····██████ beats 0/6 hard · avg 17% 0.6% typical
LiveCodeBench Pro 2025Q2 (Medium) 0.0% ●···▂··▄▂··▂ beats 2/8 very hard · avg 2.0% 0.5% sharp · too hard for it · noisy
chat & writing · #253 of 402 · 1.1% of its score
Beats [Mistral Small 4 (119B-A6.5B)](mistral-small-4.md), [Qwen2.5 Coder 32B Instruct](qwen2-5-coder-32b.md), [Ministral 3 14B Reasoning 2512](ministral-3-14b-reasoning.md) here, though they rank higher overall. Trails [Flan-T5 XXL](flan-t5-xxl.md), [Granite 4.0 H 1B](granite-4-0-h-1b.md), [Granite 4.0 1B](granite-4-0-1b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
IFEval (aggregation unspecified) 77.5% ▄▄····●··███ beats 2/8 easy · avg 78% 1.1% several runs
math · #308 of 431 · 26% of its score
Beats [Molmo 2 8B](molmo-2-8b.md), [Llama 3.1 Nemotron 70B Instruct](llama-3-1-nemotron-70b.md), [GPT-4 (0613)](gpt-4-0613.md) here, though they rank higher overall. Trails [DeepSeek R1 Distill Qwen 1.5B](deepseek-r1-distill-qwen-1-5b.md), [Gemma 7B (base)](gemma-7b-base.md), [Mistral 7B v0.1 (base)](mistral-7b-v0-1.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
HMMT February 2026 20.5% ●··▃▃·▃··▃·█ beats 1/8 hard · avg 32% 7.0% sharp · several runs
AIME 2026 29.0% ●·▄█·▄▄···█▄ beats 0/8 medium · avg 53% 6.1% sharp · at its level · several runs
AIME 2025 29.6% ●▄▄▄▄····█·▄ beats 0/8 medium · avg 42% 3.0% sharp · at its level
MATH-500 85.8% ●█▄▄▄··▄···▄ beats 1/8 very easy · avg 89% 2.2% typical
MathArena ArXivMath April 2026 (4… 1.9% ●·····▅██▄·▁ beats 0/20 very hard · avg 7.6% 1.5% sharp · too hard for it
HMMT February 2025 (maker) 22.9% ▁●▁▃▁▅▅▃██▁▄ beats 1/31 hard · avg 34% 1.2% at its level · small run
MathArena ArXivMath June 2026 (48… 4.2% ●···▄·██▆▆▄▂ beats 0/19 very hard · avg 5.1% 1.1% sharp · too hard for it
PolyMATH (maker) 26.1% ▁···●▁·▇█▄▄▄ beats 2/23 hard · avg 29% 0.9% small run
MathVista (testmini) (maker) 76.7% ▁··▁▄▄▆▇▄●█▄ beats 30/48 easy · avg 75% 0.9% at its level · small run
HMMT November 2025 (maker) 19.6% ▁●····▁▅▃█▇▁ beats 1/23 medium · avg 37% 0.8% at its level · small run
MathArena ArXivMath May 2026 (40 … 10.6% ●··▁▄█▄▁·▃·▁ beats 0/17 very hard · avg 11% 0.8% typical
DynaMath (maker) 73.6% ▂·····●··█▄▆ beats 1/10 easy · avg 71% 0.6% at its level · small run
MathArena BrokenArXiv June 2026 (… 2.8% ●··█▂▂▂▆▄··▂ beats 0/14 very hard · avg 1.3% 0.1% sharp · too hard for it · noisy
vision & documents · #120 of 153 · 14% of its score
Beats [DeepSeek V4 Flash Vision (Exp)](deepseek-v4-flash-vision-exp.md), [GPT-4 Turbo (2024-04-09)](gpt-4-turbo-2024-04-09.md), [GPT-4o mini (2024-07-18)](gpt-4o-mini-2024-07-18.md) here, though they rank higher overall.
test result vs the others in its run difficulty counts why
MMStar 61.4% ▄▄▄█·······● beats 5/5 medium · avg 61% 1.9% at its level · several runs
CharXiv descriptive questions 76.1% █·▄·······▄● beats 4/5 easy · avg 78% 1.6% at its level
ChartArena chart parsing (English… 16.2% ▁·●·▃▃▁▇▃█·▄ beats 1/22 hard · avg 27% 1.3% blunt
CharXiv (reasoning questions) (ma… 58.8% ▂▅▃▅▅●▆▅▆█▆▂ beats 43/101 medium · avg 58% 1.1% at its level · small run
MMBench-EN dev v1.1 76.0% █·····▃··▃·● beats 5/5 easy · avg 79% 1.0% blunt · several runs
MMMU (val) (maker) 64.2% ▁····▁▁▃█●▆▅ beats 28/53 medium · avg 64% 0.9% at its level · small run
MathVista (testmini) (maker) 76.7% ▁··▁▄▄▆▇▄●█▄ beats 30/48 easy · avg 75% 0.9% at its level · small run
MMMU Pro (maker) 50.3% ▂▃▁▃●▇▃▇▇██▅ beats 12/52 medium · avg 50% 0.9% at its level · small run
HallusionBench 65.5% ▄▄·▄·····●·▄ beats 4/5 easy · avg 66% 0.9% blunt · at its level · several runs
CountBench 80.5% ▄··▄··█··●·▄ beats 4/5 very easy · avg 90% 0.7% small run · several runs
OmniDocBench 1.5 overall (higher … 79.8% ▂··▂▂●··▂▂█▅ beats 3/13 easy · avg 76% 0.6% at its level · small run
RealWorldQA 69.3% ▄···█··▄▄··● beats 5/5 easy · avg 68% 0.5% blunt · at its level · several runs
SimpleVQA (maker) 38.5% ▃·●·▃·▃·█▅▃▃ beats 1/10 medium · avg 41% 0.5% blunt · at its level · small run
PMC-VQA (maker) 48.8% ▃···●··▃▃·▃█ beats 1/7 medium · avg 50% 0.5% at its level · small run
OCRBench 86.1% █··███·█···● beats 5/5 easy · avg 81% 0.4% blunt · several runs
AI2D test 75.2% █···▄●···▄·▄ beats 3/5 easy · avg 81% 0.3% blunt · at its level · several runs
CC-OCR (maker) 72.9% ▁·▂··●▂▃▃▂▆█ beats 4/26 easy · avg 73% 0.2% blunt · at its level · small run
SpaceCast-Bench predictive spatia… 31.3% ●·▅▃▅▃··▅··█ beats 0/11 very hard · avg 33% 0.1% too hard for it · noisy
ChartQA 77.5% ▄······▄··●█ beats 2/5 easy · avg 79% 0.1% blunt · noisy
PureDocBench document parsing (Av… 62.5% ▂·▁·●▂▅█·▁▂▂ beats 4/22 medium · avg 62% 0.1% blunt · at its level
web search · #10 of 15 · 0.8% of its score
Beats [Nemotron 3 Nano 4B](nemotron-3-nano-4b.md), [Gemma 4 E4B](gemma-4-e4b.md), [LFM2.5 2.6B](lfm2-5-2-6b.md) here, though they rank higher overall.
test result vs the others in its run difficulty counts why
BrowseComp-ZH 18.2% ▄▄·▄▄▄·●···█ beats 5/8 very hard · avg 13% 0.6% noisy
BrowseComp 3.1% ●···█·▃···▃▃ beats 0/7 very hard · avg 11% 0.3% noisy
other tests (39)
test result vs the others in its run difficulty counts why
MMLU-Pro 64.3% ▂··▂●·▂▂···▂ beats 2/8 medium · avg 59% 6.5% at its level · several runs
GPQA (subset unspecified) 44.6% ●···▃···▅·▃▃ beats 2/7 hard · avg 44% 3.7% sharp · several runs
MMLU-Redux 80.0% ▂···●··▂▂··▂ beats 4/8 easy · avg 76% 3.1% at its level · several runs
EuroEval Life in the UK (English … 70.6% ▂·▂●·▂·▃▃·█▃ beats 2/14 easy · avg 80% 2.7% at its level
SuperGPQA 38.6% ▃·▃·▃●█····▃ beats 4/8 medium · avg 40% 1.9% at its level · small run · several runs
MMLU 54.3% ▄·····▄●··▄█ beats 2/5 medium · avg 69% 1.7% at its level
IFBench (aggregation unspecified) 46.0% ▅····●▃·█·▃▃ beats 2/8 medium · avg 40% 1.7% at its level · small run · several runs
INCLUDE (maker) 55.4% ▂····▂▂●▄▄█▄ beats 10/36 medium · avg 59% 1.5% at its level · small run
C-Eval (maker) 73.2% ▁···▁▁▂▁▂●▃█ beats 11/33 easy · avg 75% 1.3% at its level · small run
MMMLU (maker) 63.1% ▂▁▁▂▃▂▄●▅██▃ beats 16/46 medium · avg 65% 1.3% at its level · small run
EuroEval HellaSwag (English), val… 56.2% ▂··▂●·▂▆▆█·▄ beats 2/15 medium · avg 63% 1.3% at its level
LongBench v2 24.9% ●······▅▅▃·█ beats 0/8 very hard · avg 34% 1.2% noisy · small run · several runs
Humanity's Last Exam 2.1% ●▄·▄··▄·▄·▄█ beats 0/7 very hard · avg 3.8% 1.2% too hard for it
Multi-IF (aggregation unspecified) 57.1% ▄▄···●·▄·▄██ beats 2/8 medium · avg 57% 1.1% at its level · several runs
KINA (Knowledge Index of Noah's A… 20.5% ▁▁▂●▃▄█▆▄▅▂▃ beats 4/39 very hard · avg 21% 1.1% typical
MedXPertQA MM (maker) 26.9% ▂···▂·●··▆▂█ beats 2/13 very hard · avg 28% 1.0% sharp · too hard for it · small run
MultiChallenge (maker) 33.7% ▁·▁·●▃▆▅▃▂█▅ beats 2/30 hard · avg 35% 1.0% at its level · small run
Global PIQA (maker) 69.3% ▃▃··●··▃▃▅█▅ beats 2/11 medium · avg 70% 0.9% at its level · small run
MMLU-ProX (maker) 52.3% ▁·▁●▁▂▁▃▃█▅▄ beats 2/36 medium · avg 57% 0.9% at its level · small run
MAXIFE (maker) 60.6% ▂·▂·●···▆·▄█ beats 2/11 medium · avg 63% 0.9% at its level · small run
OJBench 2.6% ●····▃▅··█▃▃ beats 0/8 very hard · avg 12% 0.8% typical
VideoMMMU (maker) 62.1% ▂···●···▂▂█▆ beats 1/10 medium · avg 63% 0.7% at its level · small run
LVBench (maker) 57.1% ▇·●▅▅▄▄··▃▁▁ beats 9/28 medium · avg 56% 0.6% at its level · small run
RefSpatial-Bench (maker) 32.9% ▃·●····▃▃▅▅█ beats 1/10 medium · avg 40% 0.6% at its level · small run
MMVU (maker) 48.6% ▃···●···▃▃▅█ beats 1/8 medium · avg 52% 0.6% at its level · small run
ERQA (maker) 43.8% ▃··●··█·▃█▅▅ beats 1/12 medium · avg 45% 0.6% at its level · small run
Video-MME (with subtitles) (maker) 75.6% ▁▁▁▁▁·●▃▁▁▁█ beats 5/16 easy · avg 75% 0.6% at its level · small run
VLMs Are Blind (maker) 75.8% ▂·●·····▃··█ beats 1/8 easy · avg 82% 0.6% small run
ZEROBench subquestions (maker) 17.1% ▄·●···▄·▄██▄ beats 1/8 hard · avg 19% 0.6% small run
MMLongBench-Doc (maker) 45.4% ▃···●···▃▅█▅ beats 1/9 medium · avg 45% 0.6% at its level · small run
SLAKE (maker) 74.4% ▄·····●▄·██▄ beats 1/7 easy · avg 73% 0.5% at its level · small run
NOVA-63 (maker) 46.4% ▃▃·●···▃█▃▃█ beats 2/11 medium · avg 47% 0.5% blunt · at its level · small run
MLVU (maker) 76.2% ▂···●···▂▄▄█ beats 1/10 easy · avg 76% 0.5% at its level · small run
EmbSpatial-Bench (maker) 77.9% ▁·····●·▁·▃█ beats 1/10 easy · avg 76% 0.5% at its level · small run
Video-MME (without subtitles) (ma… 69.0% ▁▃▁▃▁·●▄▄▅·▇ beats 8/27 medium · avg 69% 0.5% blunt · at its level · small run
MVBench (maker) 64.9% ▁··▃●▁▃▄▄▄█▄ beats 5/26 medium · avg 66% 0.2% blunt · at its level · small run
LIBRA Mini (Russian long context,… 37.4% ▄···▄●·██··▄ beats 2/8 medium · avg 38% 0.2% blunt · at its level
WMT24++ (maker) 45.8% ▁·▁●·▁·▁▂▆█▃ beats 2/27 medium · avg 62% 0.1% blunt · at its level · small run
Chess Puzzles (Epoch) 0.0% ●▁▁▁▁▂▃▃▃▂▁▁ beats 36/137 very hard · avg 5.8% 0.1% sharp · too hard for it · noisy
source results share of the score
Qwen/Qwen3.5-2B model card (submitted /… 52 37% ▇▇▇▇
openbmb/MiniCPM5-2B model card 20 26% ▇▇▇
XHToken/Spark-X2.5-4B model card 11 15% ▇▇
CohereLabs/North-Micro-Vision-Instruct … 13 9% ▇
EuroEval leaderboards 2 4% ▇
MathArena ArXivMath / BrokenArXiv / ArX… 4 3% ▇
Qwen/Qwen3.5-0.8B model card (submitted… 1 2% ▇
ChartArena: Benchmarking Chart Parsing … 1 1% ▇
Knowledge Index of Noah's Ark 1 1% ▇
ai-forever LIBRA leaderboard 1 0% ▇
SpaceCast-Bench: Evaluating Predictive … 1 0% ▇
Epoch AI 1 0% ▇
How Far Is Document Parsing from Solved… 1 0% ▇
prior: pull toward the middle 0% ▇
weighted average = 47.2
Licences: Qwen/Qwen3.5-2B model card ([apache-2.0](../../licenses/Qwen--Qwen3.5-2B.txt)) · openbmb/MiniCPM5-2B model card ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) · XHToken/Spark-X2.5-4B model card ([Apache-2.0](https://huggingface.co/XHToken/Spark-X2.5-4B/blob/0bcb35678590218655dff3765b9e61c83b35e9c4/README.md)) · CohereLabs/North-Micro-Vision-Instruct model card ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) · EuroEval leaderboards ([MIT](https://github.com/EuroEval/leaderboards/blob/1a77f1d2f8ba1ba05ab73287fb090d76bf41b2e0/LICENSE)) · MathArena ArXivMath / BrokenArXiv / ArXivLean model-output datasets ([CC BY-SA 4.0](../../licenses/matharena-sharealike-NOTICE.txt)) · Qwen/Qwen3.5-0.8B model card ([apache-2.0](../../licenses/Qwen--Qwen3.5-0.8B.txt)) · ChartArena: Benchmarking Chart Parsing across Languages ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Knowledge Index of Noah's Ark ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · ai-forever LIBRA leaderboard ([MIT](../../licenses/hf-leaderboard-spaces-MIT.txt)) · SpaceCast-Bench: Evaluating Predictive Spatial Reasoning in Vision-Language Models ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Epoch AI ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · How Far Is Document Parsing from Solved? PureDocBench: A Source-Traceable Benchmark across Clean ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run Qwen3.5 2B (Q4_K_M, needs ~1.3 GB, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Qwen3.5-2B-GGUF/tree/bb84e11355a036e28f080c7793fa6d22b7c4e344))
```sh
llama cli -hf lmstudio-community/Qwen3.5-2B-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf lmstudio-community/Qwen3.5-2B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~2 GB) `llama cli -hf lmstudio-community/Qwen3.5-2B-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run Qwen3.5 2B (Q4_K_M, needs ~1.3 GB, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Qwen3.5-2B-GGUF/tree/bb84e11355a036e28f080c7793fa6d22b7c4e344))
```sh
lms get https://huggingface.co/lmstudio-community/Qwen3.5-2B-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~2 GB) `lms get https://huggingface.co/lmstudio-community/Qwen3.5-2B-GGUF@Q8_0 -y` · 4bit (~1.7 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/Qwen3.5-2B-MLX-4bit -y` · 8bit (~2.7 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/Qwen3.5-2B-MLX-8bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run Qwen3.5 2B (Q4_K_M, needs ~1.86 GB, from [ollama.com](https://ollama.com/library/qwen3.5/tags))
```sh
ollama run qwen3.5:2b-q4_K_M
```
Other versions: Q8_0 (~3.22 GB) `ollama run qwen3.5:2b` · mxfp8 (~1.86 GB, Mac only) `ollama run qwen3.5:2b-mlx`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run Qwen3.5 2B (4bit, needs ~1.7 GB, Mac only, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Qwen3.5-2B-MLX-4bit/tree/d6e65f9fd475d078d015ee4aa3f26d683b2f1c1a))
```sh
mlx_lm.chat --model lmstudio-community/Qwen3.5-2B-MLX-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model lmstudio-community/Qwen3.5-2B-MLX-4bit
```
Other versions: 8bit (~2.7 GB, Mac only) `mlx_lm.chat --model lmstudio-community/Qwen3.5-2B-MLX-8bit`
Model repository and documentation: [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen3.5-2B)
## Published speed measurements
Reference results for the hardware, engine and settings below. These are not verified speeds of the downloads above or predictions for your machine.
hardware precision speed engine and settings
Apple M5 Pro (48GB; CPU/GPU… Q4 GGUF (4-bit;… 147.1 tok/s llama.cpp b9960, Metal backend, d…
Apple M5 Pro (48GB; CPU/GPU… MLX 4-bit (grou… 208.7 tok/s mlx-lm 0.31.3 (MLX 0.32.0), defau…
Apple M5 Pro (48GB; CPU/GPU… Q8 GGUF (8-bit;… 110.6 tok/s llama.cpp b9960, Metal backend, d…
Apple M5 Pro (48GB; CPU/GPU… MLX 8-bit (grou… 125.9 tok/s mlx-lm 0.31.3 (MLX 0.32.0), defau…
Apple M5 Max (40-core GPU),… mlx-mxfp8 174.09 tok/s ollama (version not recorded) · l…
NVIDIA GeForce RTX 3050 Ti … q8_0 30.78 tok/s ollama (version not recorded) · l…
Each speed links to its source, which has the full settings.
[interactive version](https://whatmodel.sh/model/qwen3-5-2b) · [JSON](https://whatmodel.sh/api?format=json&all=1)