---

license: apache-2.0

base_model: osunlp/QUEST-30B-RL

base_model_relation: quantized

library_name: gguf

pipeline_tag: text-generation

language:

- en

tags:

- gguf

- llama.cpp

- imatrix

- quantized

- moe

- qwen3moe

- agentic

- tool-calling

- deep-research

- intel-arc

- sycl

---

!QUEST-30B-RL · B70 Turbo

QUEST-30B-RL — B70 Turbo (Q5_K_M GGUF)

An Intel Arc / llama.cpp (SYCL) GGUF build of

osunlp/QUEST-30B-RL — imatrix-quantized to

Q5_K_M and certified to behave like the BF16 source on tool-calling and reasoning, not just

on perplexity.

| | |

|---|---|

| File | quest-30b-rl-Q5_K_M.gguf · 20.2 GB · imatrix Q5_K_M |

| Base | osunlp/QUEST-30B-RL — qwen3moe MoE, ~30.5B total / ~3.3B active, 128 experts/8 active, ctx 262144, <think> + tool-calling |

| Fits | one ~30 GiB Arc GPU at -ngl 99 (full offload) |

| Perplexity | 7.19 (wikitext-2) |

| Agentic gate vs BF16 | TOOL 8/9 · REASON 10/10 · THINK 10/10 → ✅ PASS (identical to BF16) |

| Speed (B70, measured, ship config) | prefill ~1,380 tok/s; decode ~80 tok/s short-context, ~31 tok/s @ 8K ctx |

The quant is graded against the BF16 source on tool-routing, JSON-arg integrity, multi-step reasoning,

and <think> structure — Q5_K_M scores identically to BF16 with no quant-induced regression.

A reusable certification suite (certify.sh → PASS/FAIL) ships in the pipeline repo.

Performance is real-world measured (served, end-to-end — not just llama-bench): decode holds

~80 tok/s at short context, ~31 tok/s at ~8K, and falls further with depth (it's a 3.3B-active

MoE, so fast for a 30B, but long-context decode is attention-bound); prefill is ~1,380 tok/s.

QUEST spends heavily inside <think> — budget max_tokens ≥ 1,500 for reasoning, or the final

answer can be truncated.

Run

llama-server -m quest-30b-rl-Q5_K_M.gguf -ngl 99 --port 8093 \
  -c 65536 -fa on --cache-type-k q8_0 --cache-type-v q8_0 \
  -ub 4096 -b 8192 --jinja --alias quest-30b-rl-turbo

--jinja is required (it enables <think> reasoning and tool-calling). Give agentic/reasoning

requests a real max_tokens (~4096) — the model spends tokens inside <think>.

Pipeline + standardized certification suite: github.com/newjordan/quest-turbo

---

Credits & acknowledgements

This release is a thin quantization + runtime mod on top of substantial work by others. Almost all

the value here is theirs — please cite and support them first:

• Model — OSU NLP Group (osunlp):

QUEST-30B-RL, the RL-trained deep-research / agentic

model this build simply packages. All capability is theirs. (Apache-2.0)

• Base architecture — Qwen, Alibaba: Qwen3-30B-A3B, the MoE

foundation QUEST is trained from. (Apache-2.0)

• Inference engine — llama.cpp / ggml-org: the runtime,

the GGUF format, and the k-quant / importance-matrix tooling that make this build possible.

• GPU backend — Intel oneAPI / SYCL

for Arc.

Our contribution is only the thin layer on top: imatrix calibration, the Q5_K_M quantization, an

agentic-fidelity gate, Arc/B70 SYCL speed-tuning, and packaging — by

Frosty40. Quantization scripts & docs: MIT.

Weights: Apache-2.0, inherited from the base model.

> If you use this model, please cite QUEST-30B-RL (OSU NLP) and Qwen3 (Qwen) first — this

> package would not exist without their work, or without llama.cpp.