Apodex-1.1-mini-ROCmFPX-Q8_0-AGENT-GGUF

Apodex-1.1-mini — ROCmFPX 8-bit AGENT routing for AMD Strix Halo (gfx1151)

I built this agent-tier 8-bit ROCmFPX quantization of apodex/Apodex-1.1-mini (Qwen3.5- derived hybrid MoE, 262K context, native function calling) on my Strix Halo box. The agent tier is the one I would serve for tool work.

The file

Type histogram, read from the finished file:

F32 x301, Q8_0_ROCMFPX x250, Q8_0 x182

What this build type is — and what it protects

Q8_0_ROCMFPX_AGENT (ftype 115) is my agent/tool-call coherent 8-bit routing. It keeps the LM head, token embeddings and the attention-output projections at full plain Q8_0 (182 tensors here) and routes the rest through the ROCmFPX 8-bit UE4M3-scale layout (250 tensors). My plain tier keeps zero tensors at plain Q8_0 — the routing difference is exactly what the extra ~580 MiB buys.

tie_word_embeddings is false on Apodex, so output.weight is a real standalone tensor. I verified the head types by reading the finished file back by exact tensor name (token_embd.weight and output.weight — exact match, not substring).

Text-only, trunk-only — stated up front

• No vision tower. The upstream repo is multimodal; this GGUF carries the language model only. No mmproj is included.
• No MTP head. My converter's MTP merge path dies on this checkpoint's de-fused mtp.layers.0.mlp.experts.* tensors (KeyError: 'model.layers.0.mlp.experts.0.down_proj.weight' in the Qwen2Moe merge loop), so I converted with --no-mtp. The 40-layer trunk is complete (733 tensors); speculative MTP drafting is not available from this file. I do not publish what I have not verified.

How I built it

1. Manifest gate: pulled apodex/Apodex-1.1-mini file list from the HF API with ?blobs=true and recorded the real shard bytes (15 safetensors shards, 71,903,869,048 bytes total — never the index total_size).
2. Downloaded and byte-verified all 28 files against that manifest.
3. Converted with convert_hf_to_gguf.py from my rocmfpx-dspark-halo tree (4eca07e), --outtype bf16 --no-mtp → 733 tensors, 69,376,638,528 bytes. The checkpoint ships fused expert tensors (mlp.experts.gate_up_proj / mlp.experts.down_proj); this converter's Qwen2Moe base ingests the fused layout directly.
4. Quantized with the same tree's llama-quantize at 16 threads. Dry-run estimate 34,704.57 MiB; the real file landed within ~11 MiB of it.

Measured on my box — partial-offload functional check, stated plainly

amd-halo: AMD Ryzen AI Max+ 395 (Strix Halo, gfx1151), ROCm 7.13.0, 125 GiB unified memory. At test time this box was serving 8 live llama-server seats holding ~107 GiB of unified memory, leaving me ~16 GiB. A full -ngl 999 --no-mmap load does not fit in that headroom, so I functionally checked with partial offload instead:

Greedy, port 8497, -t 16, --jinja. These t/s numbers are limited by streaming the CPU-resident layers, not by the ROCm path — treat them as load-and-generate proof with the server's own timing, not as the speed you will get on an idle box. Full-offload throughput: not measured (would require freeing the seats — I don't touch my live seats).

Sample output (greedy, prompt "Explain in one clear sentence what granite rock is primarily made of."):

<think> Thinking Process: 1. Analyze the Request: … (native thinking-mode generation)

⚠️ Stock llama.cpp will not load this file

Q8_0_ROCMFPX_AGENT is a custom tensor format that exists only in the ROCmFPX fork of llama.cpp.

llama-server -m apodex-1.1-mini-Q8_0_ROCMFPX_AGENT.gguf -dev ROCm0 -fa on -ngl 999 -c 8192   # on a box with the memory for it

Not measured

No benchmark sweeps, no context sweeps, no perplexity, no full-offload throughput — per my build discipline this is the 3-tier publish set and one functional check per tier.

Provenance & license

Converted and quantized from apodex/Apodex-1.1-mini (Apache 2.0). This quantized build is released under the same Apache 2.0 license. The ROCmFPX runtime is a third-party fork; its own terms apply to the runtime, not to these weights.

Corrected speed — full GPU offload

My first published number (0.163 tok/s) was measured with PARTIAL offload on a box already serving 8 models — that was my harness's fault, not the model's. Re-measured on an idle Ryzen AI MAX+ 395 (gfx1151, ROCm 7.2.4), full -ngl 999 offload (), 32K ctx:

32.54 tok/s (128-token generation, greedy).

<!-- VARIANTS:START -->

All my quants of Apodex-1.1-mini

All measured by me on a Ryzen AI MAX+ 395 (Strix Halo, gfx1151, ROCm 7.2.4) with the whole model on GPU (-ngl 999), 128-token greedy generation. A dash means I haven't measured that one yet — I won't put a number in a card I didn't measure.

Base model: Apodex/Apodex-1.1-mini

<!-- VARIANTS:END -->