loom-fun/openbmb-MiniCPM-OCR-1.0-loom

MiniCPM-OCR-1.0 (Q4_K_M) for AMD Loom

A Loom package (.loom programs and a model.json card) for the unmodified GGUF checkpoint openbmb/MiniCPM-OCR-1.0-gguf (MiniCPM-OCR-1_0-Q4_K_M.gguf), generated on an AMD Ryzen AI Max+ 395 "Strix Halo" with the loom-skills workflow.

Quick start

pip install "hrx-loom[hf]"
loom-fetch https://huggingface.co/loom-fun/openbmb-MiniCPM-OCR-1.0-loom/resolve/main/model.json --materialize=copy --output=minicpm-ocr-1.0
loom-serve --model=minicpm-ocr-1.0/model.json --device=amdgpu

The server listens on 127.0.0.1:8080 and answers OpenAI-style /v1/chat/completions requests.

Performance vs llama.cpp

metric	llama.cpp	Loom	margin
decode tok/s, short prompt	210.7	227.8	+8 %
decode tok/s, 2k context	203.3	208.6	+3 %
decode tok/s, 32k context	145.7	142.7	-2 %
prefill tok/s, 1k document	9679.9	10428.4	+8 %
prefill tok/s, 2k document	10173.2	10814.8	+6 %
prefill tok/s, 8k document	10279.3	10270.8	+0 %
prefill tok/s, 16k document	9043.6	8139.3	-10 %
prefill tok/s, 32k document	6660.7	5892.8	-12 %
4 streams, aggregate decode tok/s	526.3	597.9	+14 %
speculative decoding, drafted tokens accepted	n/a	n/a	n/a

• llama.cpp: build llamacpp-rocm/b1357, server arguments not recorded.
• Loom: speculative decoding off.
• Same GGUF on both engines, tokens per second. Loom: client-side timing; llama.cpp: server-side timings. Runs per value: decode and prefill not recorded (mean), 4 streams not recorded (median).
• Measured 2026-10-29; device not recorded; hrx-loom version unknown (hrx-system dd5ad00, recovered); source: loom-factory run 40835320180.
• — not measured; n/a not applicable.

Features

feature	status	detail
API	✅	OpenAI-style /v1/chat/completions
tool calling	❌	not supported
reasoning / thinking	❌	not supported
reasoning effort	❌	not supported
speculative decoding	❌	not supported
image input	❌	not supported
audio input / output	❌	not supported
sampling	❌	greedy decoding only
context size	✅	40,960 tokens per request
concurrency	✅	4 concurrent requests
max output tokens	✅	512 (preset max_tokens)

Correctness

gate	description	result
kernel_tests	Unit tests of the generated kernels against reference implementations	❌ no evidence
golden	llama.cpp reference recorded on the same GGUF: greedy tokens and top-5 log-probs, short prompts and 1k-32k documents	✅ pass ¹
compare	Greedy output vs the llama.cpp reference, token by token; every divergence is a near-tie	✅ pass ¹
vision	Image input vs llama.cpp with the same projector: identical prompt tokens, replies equal up to near-ties	— not applicable
consistency	Concurrent requests give the same output as the same requests run alone	✅ pass ¹
spec	Speculative decoding on gives exactly the output of speculative decoding off	— not applicable
agent	Multi-turn tool-calling loop: exact prompt rendering, history reuse, valid tool calls, output vs llama.cpp	— not applicable
agent_think_off	The tool-calling loop with thinking off	— not applicable
sampling	temperature, top_k and seed behave as specified; first-token distribution matches llama.cpp	❌ no evidence
needle	Needle-in-a-haystack prompt at up to half the context, compared with llama.cpp; over-capacity prompt refused	❌ no evidence
needle_full	Needle-in-a-haystack prompt at the full context is served	❌ no evidence
decision	Decision answers (choice, score, noul probabilities) vs llama.cpp on the same prompts	— not applicable

• ¹ loom-factory run 40835320180, 2026-10-29; device not recorded; hrx-loom version unknown (hrx-system dd5ad00, recovered).
• Gates are the loom-benchmark correctness gates (gates.sh); a gate without a recorded run shows no evidence.

Files

• model.json / model.json.lock: the card and the pinned source files (sha256 checked by loom-fetch).
• program/*.loom, weights.loom: the generated Loom programs.
• tokenizer.json: the tokenizer.
• llm_config.json: the configuration the programs were generated from.
• info.json: the measurements, gate evidence and versions this card shows; everything else comes from the files above.
• REPORT.md: the report written when this package was made.

Versions: card schema loom-card/2; loom-skills b0f5897 (recovered); hrx-loom version unknown (hrx-system dd5ad00, recovered); card generator cf02e66.

Source files, fetched by loom-fetch from their own repositories:

• checkpoint: openbmb/MiniCPM-OCR-1.0-gguf @ 97b10e7, MiniCPM-OCR-1_0-Q4_K_M.gguf