README.md

metadata

language:
  - en
license: apache-2.0
library_name: transformers
tags:
  - chat
  - TensorBlock
  - GGUF
pipeline_tag: text-generation
datasets:
  - anthracite-org/c2_logs_32k_llama3_qwen2_v1.2_no_system
  - anthracite-org/kalo-opus-instruct-22k-no-refusal-no-system
  - anthracite-org/kalo-opus-instruct-3k-filtered-no-system
  - anthracite-org/nopm_claude_writing_fixed
  - anthracite-org/kalo_opus_misc_240827_no_system
  - anthracite-org/kalo_misc_part2_no_system
base_model: anthracite-org/magnum-v4-12b
model-index:
  - name: magnum-v4-12b
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: HuggingFaceH4/ifeval
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 33.93
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v4-12b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: BBH
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 30.5
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v4-12b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: hendrycks/competition_math
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 9.82
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v4-12b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 6.15
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v4-12b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MuSR (0-shot)
          type: TAUR-Lab/MuSR
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 10.36
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v4-12b
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU-PRO (5-shot)
          type: TIGER-Lab/MMLU-Pro
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 28.93
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v4-12b
          name: Open LLM Leaderboard

Feedback and support: TensorBlock's Twitter/X, Telegram Group and Discord server

anthracite-org/magnum-v4-12b - GGUF

This repo contains GGUF format model files for anthracite-org/magnum-v4-12b.

The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4011.

Run them on the TensorBlock client using your local machine ↗

Prompt template

<s>[INST]{system_prompt}

{prompt}[/INST]

Model file specification

Filename	Quant type	File Size	Description
magnum-v4-12b-Q2_K.gguf	Q2_K	4.791 GB	smallest, significant quality loss - not recommended for most purposes
magnum-v4-12b-Q3_K_S.gguf	Q3_K_S	5.534 GB	very small, high quality loss
magnum-v4-12b-Q3_K_M.gguf	Q3_K_M	6.083 GB	very small, high quality loss
magnum-v4-12b-Q3_K_L.gguf	Q3_K_L	6.562 GB	small, substantial quality loss
magnum-v4-12b-Q4_0.gguf	Q4_0	7.072 GB	legacy; small, very high quality loss - prefer using Q3_K_M
magnum-v4-12b-Q4_K_S.gguf	Q4_K_S	7.120 GB	small, greater quality loss
magnum-v4-12b-Q4_K_M.gguf	Q4_K_M	7.477 GB	medium, balanced quality - recommended
magnum-v4-12b-Q5_0.gguf	Q5_0	8.519 GB	legacy; medium, balanced quality - prefer using Q4_K_M
magnum-v4-12b-Q5_K_S.gguf	Q5_K_S	8.519 GB	large, low quality loss - recommended
magnum-v4-12b-Q5_K_M.gguf	Q5_K_M	8.728 GB	large, very low quality loss - recommended
magnum-v4-12b-Q6_K.gguf	Q6_K	10.056 GB	very large, extremely low quality loss
magnum-v4-12b-Q8_0.gguf	Q8_0	13.022 GB	very large, extremely low quality loss - not recommended

Downloading instruction

Command line

Firstly, install Huggingface Client

pip install -U "huggingface_hub[cli]"

Then, downoad the individual model file the a local directory

huggingface-cli download tensorblock/magnum-v4-12b-GGUF --include "magnum-v4-12b-Q2_K.gguf" --local-dir MY_LOCAL_DIR

If you wanna download multiple model files with a pattern (e.g., *Q4_K*gguf), you can try:

huggingface-cli download tensorblock/magnum-v4-12b-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'