whatmodel.sh / WizardLM 13B V1.2

WizardLM 13B V1.2

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

WizardLM 13B V1.2 is an open-weights language model from Microsoft, released July 2023. Its published size is 13B parameters. Licence: llama2.

Catalogue identity notes

Comparison-only. Arena entry 'wizardlm-13b' (text). LMArena's leaderboard table (arena-leaderboard space @6f2eeeb) labels 'wizardlm-13b' as 'WizardLM-13b-v1.2' (Microsoft, Llama 2 Community), linking https://huggingface.co/WizardLM/WizardLM-13B-V1.2. Maker repo WizardLMTeam/WizardLM-13B-V1.2 @cf5f403: licence tag llama2. Parameters: card: 'trained from Llama-2 13b'. Released 2023-07: month the maker's Hugging Face repo was created (createdAt 2023-07-25). Arena key 'wizardlm-13b'; LMArena's table labels it 'WizardLM-13b-v1.2' linking WizardLM/WizardLM-13B-V1.2 (now WizardLMTeam/).

35.8±1.1#593 of 719 LLMsfrom 14 benchmark results across 5 sources

$ how it compares

The LLMs ranked either side of WizardLM 13B V1.2, out of 828 with a score, then more from Microsoft.

#modelscorewhereRAMreleased
685Granite 3.2 2B Instruct IBM36±1local~3GB2025-02
686Mistral 7B Instruct v0.2 Mistral36±1local~6GB2023-12
687WizardLM 13B V1.2 Microsoft36±1local~9.8GB2023-07
688Tulu 2 7B Allen Institute for AI36±2local~5.8GB2023-11
689xDAN-L1-Chat-RL-v1 xDAN-AI36±1local~6.1GB2023-12
more from Microsoft
252Phi-4-reasoning-plus 14B Microsoft59±1local~10GB2025-04
261Phi-4-reasoning 14B Microsoft58±1api $0.22/M +local~10GB2025-04
365Phi-4 14B Microsoft53±1api $0.087/M +local~10GB2024-12

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▆██▇▆▄▃▁ #365 of 458 · 4.6% of its score · 4 tests

Beats LFM2.5 8B-A1B, Qwen3-VL 2B Thinking, DeepSeek R1 Distill Qwen 7B here, though they rank higher overall. Trails Qwen2.5 1.5B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
CodeReviewQA: solution identification (hard), invariant accuracy4.6% ✎▃··▃▂▂▃▂█▃▄▂ beats 5/47medium · avg 54%▇ 1.8%sharp · noisy
CodeReviewQA: automated code refinement, exact match rate22.9% ✎▁▁▁▁▁▁▂▂▃█▅▄ beats 15/47hard · avg 33%▇ 1.6%typical
CodeReviewQA: change localisation (hard), invariant accuracy0.1% ✎█▂▁▁▂▂▂▃▅▅▅▇ beats 0/47medium · avg 36%▇ 1.2%sharp · too hard for it · noisy
CodeReviewQA: change type recognition, invariant accuracy65.1% ✎▁▁▁······▁▃█ beats 12/47easy · avg 67%▇ 0.1%blunt · at its level · noisy
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #344 of 393 · 51% of its score · 4 tests

Beats Gemma 4 E2B, Code Llama 70B Instruct, Phi-3-mini-4k-instruct (June 2024 update) here, though they rank higher overall. Trails Gemma 3 270M IT, SmolLM2 360M Instruct, Granite 4.0 H 350M here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
AlpacaEval 2 length-controlled14.5% ✎▁▂▂▃▆▆█▆▄▄▁▁ beats 35/122medium · avg 39%▇▇▇▇▇▇▇▇▇ 22%typical
Polyrating rating (bias-aware re-fit of Chatbot Arena votes, mid-2024)1083.0 ✎▂▁▂▁▃▇█▆▇▄▂▂ beats 39/95avg 1201▇▇▇▇▇ 12%blunt
LMArena (text)1149.5 ✎▁▁▁▂▃▃▃▆▆▇█▃ beats 39/379avg 1268▇▇▇▇▇ 12%blunt
AlpacaEval 1.0 win rate (vs text-davinci-003, GPT-4 judge)89.2% ✎▂▂·▆·▇▃▆█▆▄▂ beats 29/45very easy · avg 96%▇▇ 5.2%sharp · noisy
math▁▂▃▅▆▅▄█▄▁▁▁ #365 of 398 * · 7.6% of its score · 1 test

Beats Yi 34B Chat, Command R (03-2024), DeepSeek LLM 67B Chat here, though they rank higher overall. Trails Gemma 3 1B, OLMo 2 1B Instruct (0425), Stable Code 3B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
GSM8K48.6% ✎▂▂▂▅▅▆▅▇█▇▄▃ beats 43/146easy · avg 82%▇▇▇ 7.6%sharp · at its level
other tests5 not in a task · 36% of its score
testresultvs the others in its rundifficultycountswhy
BIG-Bench Hard38.5% ✎▁▁▁▁▂▄▅▅█▃▄▁ beats 22/129medium · avg 68%▇▇▇▇▇ 12%sharp
ARC Challenge50.4% ✎▆▁▁▂▂▅▇█▄▃▃▂ beats 59/97easy · avg 86%▇▇▇▇ 10%sharp · at its level
MMLU42.8% ✎▁▁▁▁▁▂▃▅█▃▆▃ beats 22/183easy · avg 74%▇▇▇ 6.5%at its level
AGIEval English38.3% ✎█▄▄··▄··▄█·█ beats 0/9easy · avg 69%▇▇ 5.8%at its level
ARC Easy69.0% ✎▁▂▃▂▃▅▇█▇▆▄▆ beats 39/65easy · avg 88%▇ 3.0%at its level

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run WizardLM 13B V1.2 · needs ~7.9 GB · from huggingface.co/TheBloke
    chat in the terminal
    llama cli -hf TheBloke/WizardLM-13B-V1.2-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf TheBloke/WizardLM-13B-V1.2-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run WizardLM 13B V1.2 · needs ~7.9 GB · from huggingface.co/TheBloke

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/TheBloke/WizardLM-13B-V1.2-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run WizardLM 13B V1.2 · needs ~7.9 GB · from huggingface.co/TheBloke
    chat in the terminal
    ollama run hf.co/TheBloke/WizardLM-13B-V1.2-GGUF:Q4_K_M

Other ways to run it

  1. Run it as published · needs ~26.03 GB · from huggingface.co/WizardLMTeam
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='WizardLMTeam/WizardLM-13B-V1.2', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Weights and documentation: huggingface.co/WizardLMTeam ↗

$ how the score was worked out · Orca 2: Teaching Small Langua… 44%, AlpacaEval 1.0 and 2.0 leader… 27%, Polyrating: A Cost-Effective… 12% and 2 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Orca 2: Teaching Small Language Models How to Reason (Mitra et al., Microsoft)6▇▇▇▇▇▇▇▇▇▇▇▇▇ 44%CC BY 4.0
AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository)2▇▇▇▇▇▇▇▇ 27%Apache-2.0
Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation (arXiv:2409.00696v3)1▇▇▇▇ 12%CC-BY-4.0
LMArena leaderboard dataset1▇▇▇▇ 12%CC BY 4.0
CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models (arXiv:2503.16167v2)4▇ 5%CC-BY-4.0
prior · pull toward the middle▇ 1%
weighted average= 35.8

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json