whatmodel.sh / Jamba Mini 1.7

Jamba Mini 1.7

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

Jamba Mini 1.7 is an open-weights language model from AI21 Labs, released July 2025. Its published size is 52B parameters (12B active per token) and a 256k-token context window. Licence: jamba-open-model-license.

Catalogue identity notes

AI21 docs deprecation table (observed 2026-10-09): 'Jamba Mini 1.7 | 2025-07 | jamba-mini-1.7-2025-08 | 2026-02-01' (API removed 2026-02-01, so api null; weights remain). Weights ai21labs/AI21-Jamba-Mini-1.7 @e8986f3 (created 2025-07-01, licence other: jamba-open-model-license, gated). Vectara row 'ai21labs/jamba-mini-1.7-2025-07' (the 2025-07 Jamba Mini 1.7 snapshot; the docs print the endpoint as jamba-mini-1.7-2025-08 — same model version 1.7).

Model information sources

Release date ↗ · Weights licence ↗

HF safetensors metadata 51,570,323,328 (BF16), same architecture size as Jamba Mini (AI21 docs: Jamba Mini '52B parameters (12B active)').

48.7±14.9#480 of 949 LLMsfrom 1 benchmark result across 1 source* thinly tested, treat the rank as provisional

$ how it compares

The LLMs ranked either side of Jamba Mini 1.7, out of 1132 with a score, then more from AI21 Labs.

#modelscorewhereRAMreleased
613Claude 3 Sonnet Anthropic49±1comparison only2024-02
614AceReason-Nemotron-1.1-7B NVIDIA49±2local~6GB2025-06
615Jamba Mini 1.7 AI21 Labs49±15local~34GB2025-07
616Hulu-Med-7B ZJU AI4H49±5comparison only2025-10
617Mixtral 8x22B Instruct v0.1 Mistral49±1local~56GB2024-04
more from AI21 Labs
332Jamba 1.6 Large AI21 Labs58±11local~839GB2025-02
485Jamba 1.6 Mini AI21 Labs53±10local~34GB2025-02
526Jamba2 Mini AI21 Labs52±15api price unknown +local~32GB2026-01

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests1 not in a task · 1.0% of its score
testresultvs the others in its rundifficultycountswhy
Vectara hallucination leaderboard (HHEM-2.3, factual consistency rate)85.3% ✎▁▂▁▃▅█▂▃▄▃▁▁ beats 11/98very easy · avg 90%▇ 1.0%blunt

$ 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 Jamba Mini 1.7 · needs ~31.2 GB · from huggingface.co/bartowski
    chat in the terminal
    llama cli -hf bartowski/ai21labs_AI21-Jamba-Mini-1.7-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf bartowski/ai21labs_AI21-Jamba-Mini-1.7-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 Jamba Mini 1.7 · needs ~31.2 GB · from huggingface.co/bartowski

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/bartowski/ai21labs_AI21-Jamba-Mini-1.7-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Jamba Mini 1.7 · needs ~31.2 GB · from huggingface.co/bartowski
    chat in the terminal
    ollama run hf.co/bartowski/ai21labs_AI21-Jamba-Mini-1.7-GGUF:Q4_K_M

Other ways to run it

  1. Run it as published · needs ~103.14 GB, NVIDIA only · from huggingface.co/ai21labs (official)
    vllm serve ai21labs/AI21-Jamba-Mini-1.7

Official weights and documentation: huggingface.co/ai21labs ↗

$ how the score was worked out · Vectara hallucination leaderb… 1%

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
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)1▇ 1%Apache-2.0
prior · pull toward the middle▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 99%
weighted average= 48.7

$ 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