Downloads · 30 days
476
65% of all-time downloads
tomvaillant/gemma4-12b-spotlight-orchestrator-v5-GGUF
gemma4-12b-spotlight-orchestrator-v5-GGUF is a text generation model from tomvaillant. Use it when you need the model to write or continue text. The card lists the license as gemma.
Q4KM quantization of the v5 Spotlight orchestrator — the servable artifact for the 12B local tier of Spotlight. It is merged from the v5 LoRA adapter with Unsloth's native merge path, coherence-checked, and converted…
Downloads · 30 days
476
65% of all-time downloads
All-time downloads
729
Public
Repo size
7.4 GB
Likes
0
Public
Click a slice to open those files.
.gguf7.4 GB · 100%
From the Hugging Face model README
Q4_K_M quantization of the v5 Spotlight orchestrator — the servable artifact for the 12B local tier of Spotlight. It is merged from the v5 LoRA adapter with Unsloth's native merge path, coherence-checked, and converted with llama.cpp.
gemma-4-12b-spotlight-orchestrator-Q4_K_M.gguf — the single 7.381 GB,
text-only Q4_K_M deployment artifact. Vision weights are intentionally not
included.llama-server -m gemma-4-12b-spotlight-orchestrator-Q4_K_M.gguf \
--port 8093 --host 127.0.0.1 \
--ctx-size 81920 --parallel 2 --no-cache-idle-slots \
--n-gpu-layers 999 --jinja --flash-attn on \
--cache-type-k q8_0 --cache-type-v q8_0 \
--reasoning-budget 400 \
--alias gemma-4-12b-spotlight-v5
These settings are part of the tested configuration:
--parallel 2 --ctx-size 81920 gives each slot 40,960 tokens, matching the
tuning context and leaving enough room for a full investigation loop.--no-cache-idle-slots prevents slot save/restore failures during subagent
delegation.--reasoning-budget 400 prevents reasoning from consuming the response
budget before a tool call or gate presentation is emitted.--jinja enables tool calling; q8 KV and flash attention keep memory use
inside the local 48 GB target.Drive Spotlight's brief → methodology → research → fact-check → report pipeline;
delegate constrained investigation and fact-check work; obey human gates; and
recover cleanly when deterministic validators reject structured inputs. The v5
dataset contains the report-execution recovery contexts that v4 fumbled, but the
production harness no longer asks any model to edit the 20 KB HTML template:
Spotlight renders all three report artifacts deterministically from validated
findings.json and fact-check.json.
Released 2026-07-10 after the preserved v4 failure fixture passed the canonical report finalizer: evidence-chain validation, deterministic Markdown/HTML/evidence map rendering, report-gate validation, and byte-identical repeat rendering. The same finalizer also passed the completed 26B case. A fresh full 12B E2E is the post-release validation; report file construction is no longer part of that model-capability bet.
curl -fsSL https://spotlight.buriedsignals.com/install-spotlight.sh | bash
Direct GGUF download:
hf download tomvaillant/gemma4-12b-spotlight-orchestrator-v5-GGUF \
gemma-4-12b-spotlight-orchestrator-Q4_K_M.gguf
The model does not replace source verification. Spotlight's deterministic fact-check validator and report finalizer are required deployment components. Tool use is tuned against Spotlight's harness schema and may not transfer to other agent frameworks without adaptation. The Q4_K_M artifact is text-only; use a separate vision-capable fact-check tier for scanned documents or imagery.