Image-Text-to-Text
GGUF
llama.cpp
amd
rocm
gfx1151
strix-halo
mixture-of-experts
mtp
long-context
vision
conversational
Instructions to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0 # Run inference directly in the terminal: llama cli -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0 # Run inference directly in the terminal: llama cli -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0 # Run inference directly in the terminal: ./llama-cli -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0 # Run inference directly in the terminal: ./build/bin/llama-cli -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Use Docker
docker model run hf.co/jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
- LM Studio
- Jan
- vLLM
How to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
- Ollama
How to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with Ollama:
ollama run hf.co/jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
- Unsloth Desktop
- Pi
How to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with Docker Model Runner:
docker model run hf.co/jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
- Lemonade
How to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Run and chat with the model
lemonade run user.Qwen3.8-Flash-CIRU-STRIX-Orca-Q8_0
List all available models
lemonade list
- Hermes Agent
How to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:8080/v1 \ --custom-model-id "jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca:Q8_0" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Check the selected runtime before model load and clarify launcher upgrades
Browse files- README.md +5 -2
- launcher-checksums.sha256 +1 -0
- run-server.sh +32 -0
- v4.3.0-checksums.sha256 +1 -1
README.md
CHANGED
|
@@ -50,11 +50,14 @@ Existing v4.2 users can update the launcher without rebuilding or downloading we
|
|
| 50 |
|
| 51 |
[Hermes task results and grading notes](benchmarks/v4.3.0/HERMES-RESULTS.md) | [v4.3 release](https://github.com/ciru-ai/Qwen3.8-Flash-CIRU-STRIX-IU4/releases/tag/v4.3.0)
|
| 52 |
|
| 53 |
-
|
|
|
|
|
|
|
| 54 |
|
| 55 |
```bash
|
| 56 |
hf download jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca \
|
| 57 |
-
--revision
|
|
|
|
| 58 |
RUNTIME_DIR="$(realpath ./ciru-runtime-v4.2.0)" bash ./model/run-server.sh
|
| 59 |
```
|
| 60 |
|
|
|
|
| 50 |
|
| 51 |
[Hermes task results and grading notes](benchmarks/v4.3.0/HERMES-RESULTS.md) | [v4.3 release](https://github.com/ciru-ai/Qwen3.8-Flash-CIRU-STRIX-IU4/releases/tag/v4.3.0)
|
| 52 |
|
| 53 |
+
**Launcher update, 16 September:** the current `main` launcher checks the HIP library selected by the dynamic loader for the v4.2 indexed-attention correction before loading the model. It stops with upgrade instructions if an older or incomplete runtime is selected, and prints the actual server/library paths and library hash. Updating the model folder alone does not upgrade `RUNTIME_DIR`, `BUILD_DIR` or `SERVER_BIN`. This check was added after the immutable v4.3.0 tag; fetch the current launcher below to use it with an existing corrected v4.2/v4.3 installation.
|
| 54 |
+
|
| 55 |
+
For an existing **corrected v4.2 or v4.3** installation, download and verify the current launcher:
|
| 56 |
|
| 57 |
```bash
|
| 58 |
hf download jcbtc/Qwen3.8-Flash-CIRU-STRIX-Orca \
|
| 59 |
+
--revision main run-server.sh launcher-checksums.sha256 --local-dir ./model
|
| 60 |
+
(cd model && sha256sum -c launcher-checksums.sha256)
|
| 61 |
RUNTIME_DIR="$(realpath ./ciru-runtime-v4.2.0)" bash ./model/run-server.sh
|
| 62 |
```
|
| 63 |
|
launcher-checksums.sha256
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
5584c936f4bfe1202a06d4a1a3891889e539ddf753f336505b2951d37a7258e5 run-server.sh
|
run-server.sh
CHANGED
|
@@ -90,6 +90,38 @@ done
|
|
| 90 |
server_bin="$(realpath "${server_bin}")"
|
| 91 |
export LD_LIBRARY_PATH="$(dirname "${server_bin}")${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}"
|
| 92 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
mkdir -p "${slot_dir}"
|
| 94 |
|
| 95 |
set -a
|
|
|
|
| 90 |
server_bin="$(realpath "${server_bin}")"
|
| 91 |
export LD_LIBRARY_PATH="$(dirname "${server_bin}")${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}"
|
| 92 |
|
| 93 |
+
# Check the library selected by the loader, including BUILD_DIR/SERVER_BIN overrides.
|
| 94 |
+
unset LD_PRELOAD
|
| 95 |
+
runtime_fail() {
|
| 96 |
+
echo "CIRU runtime check: $*" >&2
|
| 97 |
+
echo "Selected server: ${server_bin}" >&2
|
| 98 |
+
echo "Updating the model folder or launcher does not update the inference runtime." >&2
|
| 99 |
+
echo "Build/install CIRU v4.2.0 or newer, then set RUNTIME_DIR and BUILD_DIR to that installation." >&2
|
| 100 |
+
echo "Keep its executable and shared libraries together; no model-weight download is needed." >&2
|
| 101 |
+
exit 2
|
| 102 |
+
}
|
| 103 |
+
for runtime_tool in ldd grep sha256sum; do
|
| 104 |
+
command -v "${runtime_tool}" >/dev/null || runtime_fail "Missing runtime-check tool: ${runtime_tool}"
|
| 105 |
+
done
|
| 106 |
+
[[ -z "${GGML_BACKEND_PATH:-}" ]] || runtime_fail "Clear GGML_BACKEND_PATH; an extra backend can bypass this runtime check."
|
| 107 |
+
runtime_dependencies="$(ldd "${server_bin}" 2>&1)" || runtime_fail "Cannot resolve shared libraries for this server. Check the SDK and loader dependencies."
|
| 108 |
+
hip_library=""
|
| 109 |
+
while IFS= read -r runtime_line; do
|
| 110 |
+
if [[ "${runtime_line}" == *"libggml-hip.so"*" => "* ]]; then
|
| 111 |
+
hip_library="${runtime_line#* => }"
|
| 112 |
+
hip_library="${hip_library% (*}"
|
| 113 |
+
fi
|
| 114 |
+
done <<< "${runtime_dependencies}"
|
| 115 |
+
[[ -n "${hip_library}" && -f "${hip_library}" ]] || runtime_fail "The selected server does not resolve a HIP library. Use the matching CIRU shared-library build."
|
| 116 |
+
hip_library="$(realpath "${hip_library}")"
|
| 117 |
+
for correction_marker in flash_attn_index_mask_clear flash_attn_index_mask_set flash_attn_index_mask_empty; do
|
| 118 |
+
grep -aFq "${correction_marker}" "${hip_library}" || runtime_fail "The HIP library lacks the v4.2 indexed-attention correction: ${hip_library}"
|
| 119 |
+
done
|
| 120 |
+
hip_sha256="$(sha256sum "${hip_library}")"
|
| 121 |
+
hip_sha256="${hip_sha256%% *}"
|
| 122 |
+
echo "CIRU runtime check: indexed-attention correction detected; HIP SHA256 ${hip_sha256}" >&2
|
| 123 |
+
echo "CIRU runtime check: server ${server_bin}; HIP ${hip_library}" >&2
|
| 124 |
+
|
| 125 |
mkdir -p "${slot_dir}"
|
| 126 |
|
| 127 |
set -a
|
v4.3.0-checksums.sha256
CHANGED
|
@@ -1,4 +1,4 @@
|
|
| 1 |
-
|
| 2 |
e44b9ad60148bb5843070b7e89120bfa0526b0bedfcf488bbbce237e91d203f1 benchmarks/v4.3.0/HERMES-80-CASES.csv
|
| 3 |
e0a630f97621352cae224f1d07acb1a6298bca8564c89115fbbf4da336a10ec9 benchmarks/v4.3.0/HERMES-POOLED.json
|
| 4 |
b9fa142b24571b37f8a3d7bb7f271a28d800018f29fbec506cb857a02f38aa9d benchmarks/v4.3.0/HERMES-RESULTS.md
|
|
|
|
| 1 |
+
5584c936f4bfe1202a06d4a1a3891889e539ddf753f336505b2951d37a7258e5 run-server.sh
|
| 2 |
e44b9ad60148bb5843070b7e89120bfa0526b0bedfcf488bbbce237e91d203f1 benchmarks/v4.3.0/HERMES-80-CASES.csv
|
| 3 |
e0a630f97621352cae224f1d07acb1a6298bca8564c89115fbbf4da336a10ec9 benchmarks/v4.3.0/HERMES-POOLED.json
|
| 4 |
b9fa142b24571b37f8a3d7bb7f271a28d800018f29fbec506cb857a02f38aa9d benchmarks/v4.3.0/HERMES-RESULTS.md
|