whatmodel.sh / Bielik 11B v2.3 Instruct

Bielik 11B v2.3 Instruct

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

Bielik 11B v2.3 Instruct is an open-weights language model from SpeakLeash, released August 2024. Its published size is 11.17B parameters. Licence: apache-2.0.

Catalogue identity notes

Polish-focused 11B instruct model by SpeakLeash & ACK Cyfronet AGH (maker repo speakleash/Bielik-11B-v2.3-Instruct @73f5f7f, licence tag apache-2.0; repo created 2024-08-30). Appears in multi-model runs of several ledger sources: BFCL v3 (2025-06-14) and V4 (2025-12-16) as 'Bielik-11B-v2.3-Instruct (Prompt)', SpeakLeash Polish EQ-Bench and CPTUB, EuroEval. Not v2.0/v2.1/v2.2/v3.

Model information sources

Release date ↗ · Weights licence ↗ · Parameter count ↗

11,168,796,672 parameters from the Hugging Face safetensors metadata (F16; dense, so active = total).

49.8±2.2#346 of 719 LLMsfrom 6 benchmark results across 5 sources

$ how it compares

The LLMs ranked either side of Bielik 11B v2.3 Instruct, out of 828 with a score, then more from SpeakLeash.

#modelscorewhereRAMreleased
414Yi 34B 200K (base) 01.AI50±5local~23GB2023-11
415Gemma 2 27B IT Google50±1local~18GB2024-06
416Bielik 11B v2.3 Instruct SpeakLeash50±2local~8.6GB2024-08
417Qwen2.5-VL 7B Instruct Alibaba50±1local~6.4GB2025-01
418Molmo 72B (0924) Ai250±2local~309GB2024-09
more from SpeakLeash
325Bielik 11B v3.0 Instruct SpeakLeash55±2local~8.1GB2025-11
338Bielik 11B v2.6 Instruct SpeakLeash54±3local~8.1GB2025-06
371Bielik 11B v2.1 Instruct SpeakLeash53±3local~8.5GB2024-08

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #280 of 300 * · 29% of its score · 2 tests

Trails Granite 4.0 H 350M, Qwen3 0.6B, Qwen2.5 1.5B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
BFCL V4 (December 2025 archive)21.9% ✎▁▁▁▂██▅▅▆▂▂▂ beats 7/77hard · avg 32%▇▇▇▇▇▇▇▇ 19%at its level
BFCL v340.8% ✎▁▁▁▁▁▃▅█▅▄▁▁ beats 16/114medium · avg 52%▇▇▇▇ 10%blunt · at its level
other tests4 not in a task · 69% of its score
testresultvs the others in its rundifficultycountswhy
CPTUB average (SpeakLeash, 0-5 judge scale) lab72.6% ✎▁▂▂▂▃█▃▆▅▄▂▃ beats 60/95easy · avg 66%▇▇▇▇▇▇▇▇▇▇▇▇ 30%at its level
Polish EQ-Bench v2 (SpeakLeash) lab70.9% ✎▁··▁▁▂▂▆▅▆█▆ beats 54/70medium · avg 64%▇▇▇▇▇▇▇▇▇▇ 24%at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)73.1% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 202/276medium · avg 73%▇▇▇▇ 10%at its level · noisy
EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy83.8% ✎▁▁▁▁▁▁▁▂▄█▄▃ beats 91/146easy · avg 83%▇▇ 5.0%noisy

$ 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 Bielik 11B v2.3 Instruct · needs ~6.72 GB · from huggingface.co/speakleash (official)
    chat in the terminal
    llama cli -hf speakleash/Bielik-11B-v2.3-Instruct-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf speakleash/Bielik-11B-v2.3-Instruct-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 Bielik 11B v2.3 Instruct · needs ~6.72 GB · from huggingface.co/speakleash (official)

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/speakleash/Bielik-11B-v2.3-Instruct-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 Bielik 11B v2.3 Instruct · needs ~6.72 GB · from huggingface.co/speakleash (official)
    chat in the terminal
    ollama run hf.co/speakleash/Bielik-11B-v2.3-Instruct-GGUF:Q4_K_M

MLX (Mac)

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Bielik 11B v2.3 Instruct · needs ~11.9 GB, Mac only · from huggingface.co/lknik
    chat in the terminal
    mlx_lm.chat --model lknik/Bielik-11B-v2.3-Instruct-8bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model lknik/Bielik-11B-v2.3-Instruct-8bit

Official weights and documentation: huggingface.co/speakleash ↗

$ how the score was worked out · SpeakLeash CPTUB (Complex Pol… 30%, Berkeley Function Calling Lea… 29%, SpeakLeash Polish EQ-Bench le… 24% 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
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark) · lab-reported1▇▇▇▇▇▇▇▇▇ 30%MIT
Berkeley Function Calling Leaderboard result archive2▇▇▇▇▇▇▇▇▇ 29%Apache-2.0
SpeakLeash Polish EQ-Bench leaderboard · lab-reported1▇▇▇▇▇▇▇ 24%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇▇▇ 10%Apache-2.0
EuroEval leaderboards (results collection), English generative leaderboard1▇▇ 5%MIT
prior · pull toward the middle▇ 2%
weighted average= 49.8

$ 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