Skip to content
Other

Firefly-v5-agentic-multiharness

a multi-harness agentic model for coding

1 min read

This model is an adapter based on Guilherme34/Firefly-v5, turned into a full-fledged multi-harness agentic model for writing code and tool-calling. Training was carried out using Easy-Agenticify via corrective SFT (supervised fine-tuning) and Posterior-GRPO RL. It supports a wide range of popular development environments and clients, ensuring seamless integration into familiar workflows.

Key features

  • Support for multiple harnesses: pi, claude-code, opencode, codex, hermes, mini-swe-agent, aider, terminus-2, openhands, kimi-cli, goose.
  • Built-in thinking mode (enable_thinking), enabled by default.
  • Complete elimination of critical defects in tool-calling and coding after multi-stage training.
  • Use of an orthogonal LoRA Super-Adapter (rank=352, combination_type="cat") with low subspace cosine similarity.

What you can do

  • Run a local server via vLLM to serve autocomplete and test-fixing agents.
  • Connect the model to CLI utilities like Aider or Claude Code via an environment variable for code debugging.
  • Use it in open development environments to automate programming tasks.

Example

vllm serve Guilherme34/Firefly-v5-agentic-multiharness \
  --return-tokens-as-token-ids \
  --logprobs-mode processed_logprobs \
  --enable-auto-tool-choice
Read moreDetails
Base model: Guilherme34/Firefly-v5
Export mode: super_adapter
Secondary diagonal mean cosine similarity: 0.011
Datasets: SmolDataEnvs-harbor-train, SmolDataEnvs-multiharness-sft
Discuss Telegram 0

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # Guilherme34/Firefly-v5-agentic-multiharness-Super-Adapter
    # ===== Tokens: paste yours between the quotes =====
    export HF_TOKEN=""        # huggingface.co/settings/tokens, for gated models
    # ==========================================
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download Guilherme34/Firefly-v5-agentic-multiharness-Super-Adapter --local-dir ~/models/Firefly-v5-agentic-multiharness-Super-Adapter
    # The model is in ~/models. How to run it is in the model card (link above).
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of Firefly-v5-agentic-multiharness-Super-Adapter (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

Guilherme34/Firefly-v5-agentic-multiharness-Super-Adapter
Author
Guilherme34
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

00

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments