Dirk is the Qwen3.8-27B that gets straight to the point.
📊 Benchmarks — live, in progress. Measured right now and posted as each seed lands, so any arm with hatched bars is partial and will move before it settles. Honest, in the open. More plates appear here as benchmarks start.
SWE-bench-Live: Sharp Qwen3.8-27B (Dirk) and Nail (Sharp 35B-A3B MoE) vs the stock template and cloud frontier Opus 5 (high) / Sonnet 5 — same weights, Sharp is ~2x faster to a fix on the solvable band and still out-solves Opus 5 (high); median and mean shown per arm, judge-free
MMLU-Pro board — seconds per correct and accuracy across Qwen3.6-27b, Dagger, Dirk (medium), Qwen3.8-27b (medium), and Nail (35B-A3B MoE); Dirk tops accuracy at 85.3%, and the MoE Nail is quickest to a correct answer at 43s
With our Sharp chat template, MTP, and vision baked in, the model answers lean and stays on-task out of the box. No template
wrangling: download, point oMLX at it, go. If you want it to think deeper, set the effort level through chat_template_kwargs:
Levels: low, medium, high, xhigh. Omit it for Dirk's lean default (medium). Turn thinking off entirely
with "enable_thinking": false.
What it is
Base:Qwen/Qwen3.8-27B, a dense 27B vision-language model (vision preserved).
Quant: oMLX's oQ quantizer — imatrix-calibrated mixed-precision MLX, which keeps the
model's MTP head — runtimes with multi-token-prediction speculative decoding can use it for faster generation.
Template: the Sharp chat template
(v22.3.1, Qwen 3.8-aware) — froggeric's fixed Qwen template plus an always-on terseness system prompt, and turning off xhigh thinking default.
It replaces the checkpoint's chat_template.jinja; the weights and the MTP tensors are untouched.
The only thing Dirk changes versus the stock quant is the template. Same weights, asked better.
Proven on Nail and Dagger
Dirk is new, but the template is not. The identical terseness edit, measured on
Dagger's base (ThinkingCap-27B,
same weights, only the template swapped):
stock template
Sharp template
change
Claw-Eval, answer component
59.3
66.7
+7.4
Claw-Eval answer tokens
5393
2217
−59%
MMLU-Pro tokens per correct answer
1601
1248
−22%
Roughly: the same answers in a bit over half the words, with accuracy moving up. That is what Dirk
inherits on day one; its own measured numbers are in progress.
Thinking effort
Stock Qwen3.8-27B forces reasoning_effort=xhigh on every call — always-on maximum-effort
reasoning. Dirk removes that default, so it runs at the model's native medium effort: in both
the official and Unsloth templates, medium is the setting that injects no reasoning instruction
(only xhigh and low add one), and Dirk simply leaves it there. So Dirk thinks at the baseline and
answers terse, instead of being pushed to the ceiling on every request. Set reasoning_effort yourself
(low, medium, high, xhigh), per request, through chat_template_kwargs — the OpenAI-style
top-levelreasoning_effort field is dropped by llama.cpp and oMLX, so it must go there (see the JSON example above).
Sampling defaults: temperature 1.0, top_p 0.95, top_k 20.
Prefer to keep the files yourself? Download explicitly, then point --model at the local path:
bash
1hf download peculiar-ragdoll/Dirk-Qwen3.8-27B-MLX-oQ6e --local-dir Dirk-MLX
2python -m mlx_vlm.generate --model Dirk-MLX --max-tokens 512\3 --prompt "Explain what this function does."# text4python -m mlx_vlm.generate --model Dirk-MLX --max-tokens 512\5 --prompt "What is in this screenshot?" --image photo.jpg # vision
Requires mlx-vlm >= 0.6.3. MLX loaders without MTP support simply ignore the MTP tensors.
Both runtimes apply the embedded Sharp template automatically — nothing to pass.
Load it with mlx-vlm, not mlx-lm. This is a vision-language checkpoint. mlx_lm.load()
accepts it and then emits garbage tokens — we hit exactly that in testing before switching to
mlx_vlm, where the same files generate correctly. It is a loader mismatch, not a bad quant, but it
fails quietly. Single-run sanity figure on our box, mlx-vlm text-only with no MTP: ~25.5
tok/s decode — a smoke test, not a benchmark.
Pick your weapon
Qwen3.8-27B may be the new intelligence density frontier for local models that run on consumer hardware, but the already battle-tested Dagger and Nail both have their own use cases, in an arsenal that contains all three.
Nail-35B-A3B generates tokens 3–4× faster than 27B models, while still being very good at routine coding, debugging, knowledge work, and many other kinds of tasks — which means that for tasks that aren't too hard for it, it writes the unit test and regression test, and implements the feature in the time it takes 3.8-27B to get out of the gate. Reach for Nail when you need volume routine work done right and fast.
Dagger-27B is — unlike 3.8-27B — specifically tuned to minimize the number of thinking tokens while sacrificing minimal accuracy, which might still give it the advantage in speed-to-answer and multi-turn stamina under the context ceiling. Reach for Dagger when you need a session to survive 100 turns.
Dirk-27B is what you reach for when the task is genuinely hard and you want the strongest local answer without filler — accepting that Nail reaches an answer faster on work it can handle, and that a marathon session running 100 turns under the context ceiling is Dagger's domain, not Dirk's.
Dagger and Nail might still be your go-to workhorses for long and short tasks within their ability bands, due to their advantage in speed and stamina.
