HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M (Q4_K_M) for AMD Loom
A Loom package (.loom programs and a model.json card) for the unmodified GGUF checkpoint rippertnt/HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M-GGUF (hyperclovax-seed-text-instruct-1.5b-q4_k_m.gguf), for the AMD Ryzen AI Max+ 395 "Strix Halo", AMD Radeon AI PRO R9700 and AMD Radeon gfx1201 16 GB class (RX 9070 XT, RX 9070) GPUs, generated with the loom-skills workflow. Learn more on the Loom ecosystem website.
Quick start
Open your GPU and copy its commands.
AMD Ryzen AI Max+ 395 "Strix Halo" (gfx1151)
pipx install "hrx-loom[hf]==0.2.0"
loom-fetch huggingface:loom-fun-staging/rippertnt-HyperCLOVAX-SEED-Text-Instruct-1.5B-loom:chat --materialize=copy --output=HyperCLOVAX-SEED-Text-Instruct-1.5B
loom-serve --model=HyperCLOVAX-SEED-Text-Instruct-1.5B '--device=amdgpu://?completion_spin_ns=4294967295' --pool_backing=fixed --parameter_mode=preload
--materialize=copyputs copies of the fetched package files in the output directory.completion_spin_ns=4294967295enables active waiting for GPU completions (uses CPU while GPU work is in flight); the card's performance measurements use this setting.--pool_backing=fixedcommits the preset's memory pool up front instead of growing elastic backing.--parameter_mode=preloadonly changes how the weights are loaded on the host, so the performance rows were measured without it unless their job served with it.
In another terminal, send a request:
curl http://127.0.0.1:8080/v1/chat/completions \
-H 'Content-Type: application/json' \
-d '{"model": "naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B", "messages": [{"role": "user", "content": "Hello"}]}'
AMD Radeon AI PRO R9700 (gfx1201, 32 GB)
pipx install "hrx-loom[hf]==0.2.0"
loom-fetch huggingface:loom-fun-staging/rippertnt-HyperCLOVAX-SEED-Text-Instruct-1.5B-loom:chat-r9700 --materialize=copy --output=HyperCLOVAX-SEED-Text-Instruct-1.5B
loom-serve --model=HyperCLOVAX-SEED-Text-Instruct-1.5B '--device=amdgpu://?completion_spin_ns=4294967295' --pool_backing=fixed --parameter_mode=preload
--materialize=copyputs copies of the fetched package files in the output directory.completion_spin_ns=4294967295enables active waiting for GPU completions (uses CPU while GPU work is in flight); the card's performance measurements use this setting.--pool_backing=fixedcommits the preset's memory pool up front instead of growing elastic backing.--parameter_mode=preloadonly changes how the weights are loaded on the host, so the performance rows were measured without it unless their job served with it.
In another terminal, send a request:
curl http://127.0.0.1:8080/v1/chat/completions \
-H 'Content-Type: application/json' \
-d '{"model": "naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B", "messages": [{"role": "user", "content": "Hello"}]}'
Radeon gfx1201, 16 GB (RX 9070 XT, RX 9070)
pipx install "hrx-loom[hf]==0.2.0"
loom-fetch huggingface:loom-fun-staging/rippertnt-HyperCLOVAX-SEED-Text-Instruct-1.5B-loom:chat-16gb --materialize=copy --output=HyperCLOVAX-SEED-Text-Instruct-1.5B
loom-serve --model=HyperCLOVAX-SEED-Text-Instruct-1.5B '--device=amdgpu://?completion_spin_ns=4294967295' --memory_bytes=15000000000 --state_slab_bytes=4mib --weight_slab_bytes=64mib --parameter_mode=preload
--materialize=copyputs copies of the fetched package files in the output directory.completion_spin_ns=4294967295enables active waiting for GPU completions (uses CPU while GPU work is in flight); the card's performance measurements use this setting.--memory_bytessets the serving memory budget for this GPU.--state_slab_bytessets the allocation slab size for state memory.--weight_slab_bytessets the allocation slab size for weight memory.--parameter_mode=preloadonly changes how the weights are loaded on the host, so the performance rows were measured without it unless their job served with it.
In another terminal, send a request:
curl http://127.0.0.1:8080/v1/chat/completions \
-H 'Content-Type: application/json' \
-d '{"model": "naver-hyperclovax/HyperCLOVAX-SEED-Text-Instruct-1.5B", "messages": [{"role": "user", "content": "Hello"}]}'
The server listens on 127.0.0.1:8080 and answers OpenAI-style /v1/chat/completions requests.
Performance vs llama.cpp
Strix Halo (gfx1151)
| metric | llama.cpp | Loom | margin vs best llama.cpp | roofline | Loom % of roofline | llama.cpp Vulkan |
|---|---|---|---|---|---|---|
| decode tok/s, short prompt | 155.6 | 168.8 | -2 % | 215.5 | 78 % | 171.9 |
| decode tok/s, 2k context | 134.0 | 142.5 | -1 % | 180.1 | 79 % | 143.3 |
| decode tok/s, 32k context | 49.8 | 50.6 | +2 % | 52.1 | 97 % | 47.6 |
| prefill tok/s, 1k document | 5377.2 | 7422.9 | +38 % | 20950.1 | 35 % | 4154.3 |
| prefill tok/s, 2k document | 5389.1 | 7088.6 | +32 % | 20243.5 | 35 % | 3960.5 |
| prefill tok/s, 8k document | 4493.7 | 6385.0 | +42 % | 16836.3 | 38 % | 3013.2 |
| prefill tok/s, 16k document | 3472.5 | 4920.1 | +42 % | 13750.4 | 36 % | 2146.3 |
| prefill tok/s, 32k document | 2307.5 | 3467.9 | +50 % | 10062.0 | 34 % | 1420.8 |
| 4 streams, aggregate decode tok/s | 434.4 | 587.6 | +25 % | 697.7 | 84 % | 471.9 |
| speculative decoding, drafted tokens accepted | n/a | n/a | n/a | n/a | n/a | n/a |
Methodology
- Compared: Loom against llama.cpp ROCm
llamacpp-rocm/b1374, llama.cpp Vulkanllamacpp-vulkan/b12211; the same GGUF, the same DevLab runner type, different jobs, the same prompts; Loom served with the active wait. Values are the median of 6 sessions (a session's value is the median of its samples: 3 samples for decode_2k, decode_32k, prefill_8k; one sample for the others) (4 streams: median of 12); Loom is timed by the client, llama.cpp by its server. - llama.cpp measured on 2026-11-13 (DevLab Dispatch run 43390795404), Loom on 2026-11-17 (DevLab Dispatch run 44088689861): same runner type, different jobs (llama.cpp was not re-measured for this package).
- Out of the box on both engines: Loom serves the preset shipped in this package; llama.cpp runs its build defaults plus only the server arguments listed below (offload, context and slots). No tuning sweep on either engine.
- Drafting, 1-stream rows (decode, decode_2k, decode_32k, prefill): no drafting on either engine (no MTP head in the GGUF and no drafter published for this exact model). Each engine is loaded once for all these rows.
- 4 streams: both engines loaded with drafting off, 4 parallel requests, aggregate tokens per second.
- Rows: decode = short-context decode; decode_2k / decode_32k = decode after a ~2k / ~32k-token context; prefill_Nk = prompt processing of an N k-token prompt; streams4 = 4-stream aggregate decode. A margin within +-3 % (the measured same-runner noise) is a tie.
- llama.cpp ROCm server arguments:
1-stream rows: -ngl 99 -cram 0 -c 65536 -np 1; 4 streams: -ngl 99 -cram 0 -c 65536 -kvu -np 4. - llama.cpp Vulkan server arguments:
1-stream rows: -ngl 99 -cram 0 -c 36864 -np 1; 4 streams: -ngl 99 -cram 0 -c 147456 -np 4. - Measured on AMD DevLab Dispatch Strix Halo runners (
devlab-perfjobs); runners differ between sessions. - Margin vs best llama.cpp: (Loom / max(ROCm, Vulkan) - 1) Γ 100 %, using the available measured tokens/s for each row on this GPU. With only one valid backend, that backend is used; with neither, the margin is not measured. Raw backend values and their measurement sources are retained below and in the table.
- Measured 2026-11-17; device AMD Strix Halo (DevLab Dispatch runner X2-LA03-S01-D05), Linux; hrx-loom
0.2.0(hrx-system0cd370f, recovered); source: DevLab Dispatch run 44088689861. - llama.cpp Vulkan: measured 2026-11-13; source: DevLab Dispatch run 43390795404. Archive sha256
9759acd. - Roofline: the hardware ceiling for this GGUF on AMD Ryzen AI Max+ 395 "Strix Halo" (gfx1151) at 216 GB/s and 59 TFLOP/s, from loom-roofline
roofline.py(loom-skillsa1745dc). - β not measured; n/a not applicable.
Radeon AI PRO R9700 (gfx1201, 32 GB)
| metric | llama.cpp | Loom | margin vs best llama.cpp | roofline | Loom % of roofline | llama.cpp Vulkan |
|---|---|---|---|---|---|---|
| decode tok/s, short prompt | 297.7 | 375.9 | +9 % | 611.5 | 61 % | 344.0 |
| decode tok/s, 2k context | 262.5 | 302.6 | -1 % | 511.2 | 59 % | 306.6 |
| decode tok/s, 32k context | 110.4 | 122.6 | -2 % | 147.8 | 83 % | 125.0 |
| prefill tok/s, 1k document | 16946.6 | 22595.3 | +33 % | 75988.6 | 30 % | 12428.3 |
| prefill tok/s, 2k document | 17267.5 | 24221.3 | +40 % | 73425.5 | 33 % | 13189.6 |
| prefill tok/s, 8k document | 14521.4 | 21563.0 | +48 % | 61067.1 | 35 % | 10960.4 |
| prefill tok/s, 16k document | 11471.7 | 16970.3 | +48 % | 49874.5 | 34 % | 8714.8 |
| prefill tok/s, 32k document | 8077.5 | 11908.3 | +47 % | 36496.1 | 33 % | 6112.3 |
| 4 streams, aggregate decode tok/s | 745.2 | 1350.8 | +27 % | 1979.9 | 68 % | 1063.6 |
| speculative decoding, drafted tokens accepted | n/a | n/a | n/a | n/a | n/a | n/a |
Methodology
- Compared: Loom against llama.cpp ROCm
llamacpp-rocm/b1375, llama.cpp Vulkanllamacpp-vulkan/b12211; the same GGUF, the same DevLab runner type, different jobs, the same prompts; Loom served with the active wait. Values are the median of 5β6 sessions (a session's value is the median of its samples: 3 samples for decode_2k, decode_32k, prefill_8k; one sample for the others) (4 streams: median of 12); Loom is timed by the client, llama.cpp by its server. - llama.cpp measured on 2026-11-15 (DevLab Dispatch run 43655173096), Loom on 2026-11-17 (DevLab Dispatch run 43993245028): same runner type, different jobs (llama.cpp was not re-measured for this package).
- Out of the box on both engines: Loom serves the preset shipped in this package; llama.cpp runs its build defaults plus only the server arguments listed below (offload, context and slots). No tuning sweep on either engine.
- Drafting, 1-stream rows (decode, decode_2k, decode_32k, prefill): no drafting on either engine (no MTP head in the GGUF and no drafter published for this exact model). Each engine is loaded once for all these rows.
- 4 streams: both engines loaded with drafting off, 4 parallel requests, aggregate tokens per second.
- Rows: decode = short-context decode; decode_2k / decode_32k = decode after a ~2k / ~32k-token context; prefill_Nk = prompt processing of an N k-token prompt; streams4 = 4-stream aggregate decode. A margin within +-3 % (the measured same-runner noise) is a tie.
- llama.cpp ROCm server arguments:
-ngl 99 -cram 0 -c 65536 -np 1. - llama.cpp Vulkan server arguments:
-ngl 99 -cram 0 -c 36864 -np 1. - Measured on AMD DevLab Dispatch R9700 runners (
devlab-perfjobs); runners differ between sessions. - Margin vs best llama.cpp: (Loom / max(ROCm, Vulkan) - 1) Γ 100 %, using the available measured tokens/s for each row on this GPU. With only one valid backend, that backend is used; with neither, the margin is not measured. Raw backend values and their measurement sources are retained below and in the table.
- Measured 2026-11-17; device DevLab Dispatch runner ephemeral-devlab-sm-01-02-r9700s-4-133738799824, AMD Radeon AI PRO R9700S (gfx1201), Linux; hrx-loom
0.2.0(hrx-system commit unknown, recovered); source: DevLab Dispatch run 43993245028. - llama.cpp Vulkan: measured 2026-11-15; source: DevLab Dispatch run 43655173096. Archive sha256
9759acd. - Roofline: the hardware ceiling for this GGUF on AMD Radeon AI PRO R9700S (gfx1201, 32 GB, 2.35 GHz) at 613 GB/s and 214 TFLOP/s, from loom-roofline
roofline.py(loom-skills203cf33). - β not measured; n/a not applicable.
Radeon RX 9070 XT
| metric | llama.cpp | Loom | margin vs best llama.cpp | roofline | Loom % of roofline | llama.cpp Vulkan |
|---|---|---|---|---|---|---|
| decode tok/s, short prompt | 283.2 | 363.1 | +3 % | 638.4 | 57 % | 353.0 |
| decode tok/s, 2k context | 253.2 | 293.7 | -6 % | 533.7 | 55 % | 313.8 |
| decode tok/s, 32k context | 114.3 | 119.2 | -6 % | 154.3 | 77 % | 126.5 |
| prefill tok/s, 1k document | 17152.9 | 21581.6 | +26 % | 69099.9 | 31 % | 15445.9 |
| prefill tok/s, 2k document | 17365.2 | 23634.1 | +36 % | 66769.2 | 35 % | 16055.1 |
| prefill tok/s, 8k document | 14778.8 | 19798.2 | +34 % | 55531.1 | 36 % | 13480.5 |
| prefill tok/s, 16k document | 11219.2 | 14442.7 | +29 % | 45353.1 | 32 % | 10676.0 |
| prefill tok/s, 32k document | 7814.4 | 11522.8 | +47 % | 33187.6 | 35 % | 7353.7 |
| 4 streams, aggregate decode tok/s | 748.4 | 1333.5 | +24 % | 2067.1 | 65 % | 1078.5 |
| speculative decoding, drafted tokens accepted | n/a | n/a | n/a | n/a | n/a | n/a |
Methodology
- Compared: Loom against llama.cpp ROCm
llamacpp-rocm/b1375, llama.cpp Vulkanllamacpp-vulkan/b12211; the same GGUF, the same machine and session, the same prompts; Loom served with the active wait. Values are the median of 6 sessions (a session's value is the median of its samples: 3 samples for decode_2k, decode_32k, prefill_8k; one sample for the others) (4 streams: median of 3); Loom is timed by the client, llama.cpp by its server. - Out of the box on both engines: Loom serves the preset shipped in this package; llama.cpp runs its build defaults plus only the server arguments listed below (offload, context and slots). No tuning sweep on either engine.
- Drafting, 1-stream rows (decode, decode_2k, decode_32k, prefill): no drafting on either engine (no MTP head in the GGUF and no drafter published for this exact model). Each engine is loaded once for all these rows.
- 4 streams: both engines loaded with drafting off, 4 parallel requests, aggregate tokens per second.
- Rows: decode = short-context decode; decode_2k / decode_32k = decode after a ~2k / ~32k-token context; prefill_Nk = prompt processing of an N k-token prompt; streams4 = 4-stream aggregate decode. A margin within +-3 % (the measured same-runner noise) is a tie.
- llama.cpp ROCm server arguments:
-ngl 99 -cram 0 -c 65536 -np 1. - llama.cpp Vulkan server arguments:
-ngl 99 -cram 0 -c 36864 -np 1. - Measured on AMD Ryzen 7 9800X3D 8-Core Processor, AMD Radeon RX 9070 XT, Ubuntu 26.04, a named machine (this GPU has no DevLab runner); both engines ran on it.
- Margin vs best llama.cpp: (Loom / max(ROCm, Vulkan) - 1) Γ 100 %, using the available measured tokens/s for each row on this GPU. With only one valid backend, that backend is used; with neither, the margin is not measured. Raw backend values and their measurement sources are retained below and in the table.
- Measured 2026-11-18; device AMD Ryzen 7 9800X3D 8-Core Processor, AMD Radeon RX 9070 XT, Ubuntu 26.04; hrx-loom
0.2.0(hrx-system0cd370f); source: named-machine run. - llama.cpp Vulkan: measured 2026-11-18; source: named-machine run. Archive sha256
9759acd. - Roofline: the hardware ceiling for this GGUF on AMD Radeon RX 9070 XT (gfx1201, 16 GB) at 640 GB/s and 195 TFLOP/s, from loom-roofline
roofline.py(loom-skills6e9637a). - β not measured; n/a not applicable.
Features
| feature | status | detail |
|---|---|---|
| API | β | OpenAI-style /v1/chat/completions |
| tool calling | β | not applicable |
| reasoning / thinking | β | not applicable |
| reasoning effort | β | not applicable |
| speculative decoding | β | not applicable |
| image input | β | not applicable |
| audio input / output | β | not applicable |
| sampling | β | temperature, top_p, top_k, seed |
| context size | β | 131,072 tokens per request |
| concurrency | β | 4 concurrent requests |
| max output tokens | β | 4,096 (preset max_tokens) |
Correctness
| gate | description | Strix Halo | Radeon AI PRO R9700 | Radeon RX 9070 XT |
|---|---|---|---|---|
kernel_tests |
Unit tests of the generated kernels against reference implementations | β pass (56/56) ΒΉ | β pass (56/56) Β² | β pass (56/56) Β³ |
golden |
llama.cpp reference recorded on the same GGUF: greedy tokens and top-5 log-probs, short prompts and 1k-32k documents | β pass β΄ | β pass β΅ | β pass βΆ |
compare |
Greedy output vs the llama.cpp reference, token by token; every divergence is a near-tie | β pass β΄ | β pass β΅ | β pass βΆ |
vision |
Image input vs llama.cpp with the same projector: identical prompt tokens, replies equal up to near-ties | β not applicable | β not applicable | β not applicable |
consistency |
Concurrent requests give the same output as the same requests run alone | β pass β΄ | β pass β΅ | β pass βΆ |
spec |
Speculative decoding on gives exactly the output of speculative decoding off | β not applicable | β not applicable | β not applicable |
agent |
Multi-turn tool-calling loop: exact prompt rendering, history reuse, valid tool calls, output vs llama.cpp | β not applicable | β not applicable | β not applicable |
agent_think_off |
The tool-calling loop with thinking off | β not applicable | β not applicable | β not applicable |
sampling |
temperature, top_k and seed behave as specified; first-token distribution matches llama.cpp | β pass β΄ | β pass β΅ | β pass βΆ |
logits |
Host logits readback (llm.logits on): at T=0 every returned row's argmax is the device's token, the top-k matches llama.cpp, greedy text is unchanged | ? not re-run on this package (the last pass was on a previous build, sha256 c368639; the nightly gates re-run it) |
? not re-run on this package (the last pass was on a previous build, sha256 c368639; the nightly gates re-run it) |
β no evidence |
decision |
Decision answers (choice, score, noul probabilities) vs llama.cpp on the same prompts | β not applicable | β not applicable | β not applicable |
- ΒΉ DevLab Dispatch run 44088689861, 2026-11-17; device AMD Strix Halo (DevLab Dispatch runner X2-LA03-S01-D05), Linux; hrx-loom
0.2.0(hrx-system0cd370f, recovered). - Β² DevLab Dispatch run 43993245028, 2026-11-17; device DevLab Dispatch runner ephemeral-devlab-sm-01-02-r9700s-4-133738799824, AMD Radeon AI PRO R9700S (gfx1201), Linux; hrx-loom
0.2.0(hrx-system commit unknown, recovered). - Β³ named-machine run, 2026-11-17; device AMD Ryzen 7 9800X3D 8-Core Processor, AMD Radeon RX 9070 XT, Ubuntu 26.04; hrx-loom
0.2.0(hrx-system0cd370f). - β΄ DevLab Dispatch run 44088689861, LEVEL=dev, 2026-11-17; device AMD Strix Halo (DevLab Dispatch runner X2-LA03-S01-D05), Linux; hrx-loom
0.2.0(hrx-system0cd370f, recovered). - β΅ DevLab Dispatch run 43993245028, LEVEL=dev, 2026-11-17; device DevLab Dispatch runner ephemeral-devlab-sm-01-02-r9700s-4-133738799824, AMD Radeon AI PRO R9700S (gfx1201), Linux; hrx-loom
0.2.0(hrx-system commit unknown, recovered). - βΆ named-machine run, LEVEL=dev, 2026-11-17; device AMD Ryzen 7 9800X3D 8-Core Processor, AMD Radeon RX 9070 XT, Ubuntu 26.04; hrx-loom
0.2.0(hrx-system0cd370f). - 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 byloom-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: how the values above were measured: setup, commands and the value of every run.
Versions: card schema loom-card/3; loom-skills unknown; hrx-loom unknown; card generator b931191.
Source files, fetched by loom-fetch from their own repositories:
- checkpoint: rippertnt/HyperCLOVAX-SEED-Text-Instruct-1.5B-Q4_K_M-GGUF @
3d2edd5,hyperclovax-seed-text-instruct-1.5b-q4_k_m.gguf