Tess-4-27B-ROCmFPX-GGUF

⚠️ These files do NOT load on standard llama.cpp

They use AMD-native *_ROCMFPX tensor types from the experimental ciru-ai/ROCmFPX llama.cpp fork (build from source).

Derivative of Tess-4-27B, quantized to AMD-native ROCmFPX formats (fork-only) tuned for Strix Halo (gfx1151).

Base Model

This is a derivative of Tess-4-27B. All credit for the base model architecture and weights goes to the original authors. The base model's license applies to this derivative.

ROCmFPX (AMD-native, fork-only)

These GGUFs use AMD-native quantization schemes from the experimental ciru-ai/ROCmFPX llama.cpp fork, tuned for and benchmarked on AMD Strix Halo (Radeon 8060S iGPU, gfx1151, unified memory):

• ROCmFP3/4/6/8 tensor types with straight and "agent" presets (agent presets keep tool-calling / JSON-structured output reliable at low bit-widths)
• Files load only on the fork -- build it from source. Known-good commit these files were built and validated with:

git clone https://github.com/ciru-ai/ROCmFPX && cd ROCmFPX
git checkout 221402af8574faf652b101b6afe225a3f329561f

GGUF Files

Usage

Requires a from-source build of the ROCmFPX fork (stock llama.cpp, LM Studio, and Ollama cannot load these files):

# Interactive chat
llama-cli -m Tess-4-27B-Q3_0_ROCMFPX.gguf -c 8192 -cnv

# Server mode
llama-server -m Tess-4-27B-Q3_0_ROCMFPX.gguf -c 8192 --port 8080 -ngl 99 -fa on

Serving: MTP Speculative Decoding

This model includes MTP ("nextn") draft tensors, enabling self-speculative decoding -- measured ~1.6-1.9x faster generation with a ~95% first-token accept rate (no separate draft model needed; it drafts from itself):

llama-server -m Tess-4-27B-Q3_0_ROCMFPX.gguf -c 8192 --port 8080 --host 127.0.0.1 -ngl 99 -md Tess-4-27B-Q3_0_ROCMFPX.gguf --spec-type draft-mtp -ctk q8_0 -ctv q8_0 -fa on

Memory cost: MTP needs its own draft context alongside the main context, so serving with it uses roughly 2x the model's memory compared to serving without -md/--spec-type draft-mtp.

Caveats

• The base model's license (apache-2.0) applies to all derivative files
• Fork-only files: stock llama.cpp, LM Studio, and Ollama cannot load these -- build ciru-ai/ROCmFPX from source
• Quantization reduces precision -- verify outputs for your specific use case

Limitations

• Quantized models may exhibit subtle differences from the full-precision fine-tune
• This model inherits any limitations and biases present in the base model

---

Generated with Foundry