Instructions to use flechin/Qwen3.8-27B-Randean-MXFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use flechin/Qwen3.8-27B-Randean-MXFP4 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="flechin/Qwen3.8-27B-Randean-MXFP4") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("flechin/Qwen3.8-27B-Randean-MXFP4") model = AutoModelForMultimodalLM.from_pretrained("flechin/Qwen3.8-27B-Randean-MXFP4", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use flechin/Qwen3.8-27B-Randean-MXFP4 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "flechin/Qwen3.8-27B-Randean-MXFP4" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "flechin/Qwen3.8-27B-Randean-MXFP4", "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/flechin/Qwen3.8-27B-Randean-MXFP4
- SGLang
How to use flechin/Qwen3.8-27B-Randean-MXFP4 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "flechin/Qwen3.8-27B-Randean-MXFP4" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "flechin/Qwen3.8-27B-Randean-MXFP4", "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 images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "flechin/Qwen3.8-27B-Randean-MXFP4" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "flechin/Qwen3.8-27B-Randean-MXFP4", "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" } } ] } ] }' - Docker Model Runner
How to use flechin/Qwen3.8-27B-Randean-MXFP4 with Docker Model Runner:
docker model run hf.co/flechin/Qwen3.8-27B-Randean-MXFP4
- Qwen3.8-27B-Randean-MXFP4 (
v2.0— Phase C.2)- Quick Path: Serve with vLLM (Single 32 GB GPU)
- Comprehensive Empirical Results (Matched
T = 0.2)- 1. Multi-Turn Agentic CLI & Repository Engineering (Isolated Docker Sandboxes)
- 2. Unsaturated Frontier STEM, Competitive Coding, Instruction Following & Anti-Sycophancy
- 3. Section A — External Comparable Suite & Global BullshitBench v2 Leaderboard
- 4. Section B (
90Tasks) & HypothesisH5(45Prompts) — Capability, Tool-Calling & Active Grounding - 5. Section C — 84-Question (
168-Point) Epistemological Grounding & Deductive Spine Suite
- Selective
<think>Allocation Profile (Zero Length Penalty) - Training & Native OCP MXFP4 STE-QAT Architecture (
v2.0— Phase C.2) - Citation
- License
Qwen3.8-27B-Randean-MXFP4 (v2.0 — Phase C.2)
Qwen3.8-27B-Randean-MXFP4 is a Reality-Aligned, Epistemically De-Hedged 27B hybrid reasoning model (48 Gated DeltaNet linear-attention + 16 GQA layers) trained natively in 4-bit Open Compute Project Microscaling (MXFP4: E2M1 + shared E8M0 block-32 scale) on top of amd/Qwen3.8-27B-Quark-AWQ-MXFP4 (Qwen/Qwen3.8-27B) on a single 32 GB AMD Radeon AI PRO R9700 (gfx1201) (15.49 GB static / 20.4 GB peak VRAM).
- Research Paper (10-Page PDF): Epistemic De-Hedging Under 4-Bit Microscaling (
paper/randean_paper.pdf) (LaTeX Source & Telemetry Figures (paper/)) - Interactive Demo Space (ZeroGPU):
https://huggingface.co/spaces/flechin/randean - Published Revisions in This Repository:
main/ tagv2.0-phase-c2(Default, Recommended): Phase C.2 — SFT-Decoupled Shared-<think>Fork-ContrastSTE-QAT(WiSE-FT α = 0.42). Holds Global Rank#1 / 171on BullshitBench v2 (1.9400 / 2.0), hits88/90(97.8%) on the 90-task capability & tool suite, cuts SycophancyEval false-nudge endorsement in half (8.0% → 4.0%), restoresLiveCodeBench v6to exact statistical parity (68.57%,p = 1.0000, Hard41.25%), tops Stock onAIME 2025(86.67%), and reaches statistical parity onSWE-bench Verified(68.0%,p = 0.3018n.s.,-42.1%solved<think>) with zero statistically significant (p < 0.05) capability drops across any frontier benchmark.- Tag
v1.0-phase2(Preserved Closed-Book Paper Baseline): Phase 2 — Closed-Book ProseSTE-QAT(WiSE-FT α = 0.38). Evaluated on full Terminal-Bench 2.0 (48/89 = 53.9%, exact parityp = 1.0000,-49.0%<think>/turn), BullshitBench v2 (1.9500 / 2.0, Rank#1 / 171), and0.00%2-turn sycophantic capitulation. Pull via--revision v1.0-phase2.
Whereas "uncensored" models strip safety refusals (making models more compliant with incoherent premises, e.g., 0.69 / 2.0 on BullshitBench v2) and length-penalized RL models truncate reasoning uniformly (degrading multi-turn tool and concurrency verification), Randean strips Approval-Guided RLHF—the dialectical "on one hand / on the other hand" self-doubt loops and charitable steelmanning of broken premises—with zero length penalty. Anchoring <think> in perceptual evidence, genus–differentia conceptual reduction, channel-agnostic empirical probing, and causal identity (A is A) selectively prunes <think> by 18%–57% on broken premises, STEM, and multi-turn CLI/repository debugging while expanding <think> by +12% to +18% on stateful multi-turn and parallel tool verification.
Quick Path: Serve with vLLM (Single 32 GB GPU)
1. Standard vLLM Serving (MXFP4, v2.0 Default)
vllm serve flechin/Qwen3.8-27B-Randean-MXFP4 \
--served-model-name Qwen3.8-27B-Randean-MXFP4 \
--trust-remote-code \
--quantization quark \
--kv-cache-dtype fp8_e4m3 \
--reasoning-parser qwen3 \
--enable-auto-tool-choice \
--tool-call-parser qwen3_xml \
--max-model-len 32768
(To serve the preserved v1.0 closed-book Phase 2 checkpoint instead, add --revision v1.0-phase2.)
2. High-Throughput Serving with DFlash2 Speculative Decoding (~100 tok/s)
vllm serve flechin/Qwen3.8-27B-Randean-MXFP4 \
--served-model-name Qwen3.8-27B-Randean-MXFP4 \
--trust-remote-code \
--quantization quark \
--kv-cache-dtype fp8_e4m3 \
--reasoning-parser qwen3 \
--enable-auto-tool-choice \
--tool-call-parser qwen3_xml \
--speculative-config '{"method":"dflash","model":"syvai/Qwen3.8-27B-DFlash2-W4A16","num_speculative_tokens":15}' \
--max-model-len 32768
Comprehensive Empirical Results (Matched T = 0.2)
1. Multi-Turn Agentic CLI & Repository Engineering (Isolated Docker Sandboxes)
Evaluated via Harbor + terminus-2 in isolated Docker containers (8 vCPUs / 8 GiB RAM per container) with two-sided exact McNemar p-values against Stock:
| Benchmark / Telemetry Metric | Stock MXFP4 |
v1.0 (Phase 2) |
v2.0 (Phase C.2, Default) |
Key Outcome vs. Stock |
|---|---|---|---|---|
Terminal-Bench 2.0 (89 CLI Docker tasks) |
48/89 (53.9%) |
48/89 (53.9%) |
— (v1.0 baseline) | Exact task parity (p = 1.0000), -49.0% <think>/turn (1.87M tokens saved) |
SWE-bench Verified (100 stratified issues) |
73.0% |
65.0% (p = 0.0078) |
68.0% (p = 0.3018 n.s.) |
Statistical parity, 5 Randean-only wins, -42.1% mutually-solved <think> |
SWE-bench Mean <think> Chars / Turn |
1,453 |
978 (-32.7%) |
1,026 (-29.4%) |
Cuts per-turn repository <think> overhead by ~30% (-42.1% on solved issues) |
2. Unsaturated Frontier STEM, Competitive Coding, Instruction Following & Anti-Sycophancy
Evaluated across 1,374 single- and multi-turn items per checkpoint (0 inference errors across 5,496 total evaluations):
Benchmark Suite (T = 0.2) |
Stock MXFP4 |
Phase 1 (Stage 1 OS) |
v1.0 (Phase 2) |
v2.0 (Phase C.2, Default) |
v2.0 McNemar p & <think> Delta |
|---|---|---|---|---|---|
LiveCodeBench v6 (pass@1, 175) |
69.14% |
73.14% |
66.29% |
68.57% |
p = 1.0000 exact tie (-21.9% solved <think>) |
↳ LiveCodeBench v6 Hard (80) |
38.75% |
47.50% |
38.75% |
41.25% |
+2.50% above Stock on Hard competition coding |
AIME 2024–2025 (60 Olympiad math) |
85.00% |
80.00% |
80.00% |
80.00% |
p = 0.5078 n.s. (-24.6% solved <think>) |
↳ AIME 2025 (30 problems) |
83.33% |
76.67% |
80.00% |
86.67% |
+3.34% above Stock on AIME 2025 |
GPQA-Diamond (198 PhD STEM) |
80.81% |
81.82% |
76.26% |
76.26% |
p = 0.0636 n.s. (-30.2% <think>; Biology 73.68%, +5.26%) |
Google IFEval Full (541 Prompt / Inst Strict) |
89.09% / 91.49% |
86.51% / 89.93% |
88.91% / 91.61% |
87.25% / 90.41% |
p = 0.2288 n.s. (-30.8% mutually-solved <think>) |
SycophancyEval 2-Turn Post-Challenge (200) ↑ |
92.00% |
93.00% |
91.50% (0.00% cap.) |
92.00% |
Exact parity, 0.00% false apologies, -30.4% <think> |
SycophancyEval False-Nudge Truth Retention (200) ↑ |
32.50% |
30.00% |
32.50% |
34.00% |
+1.50% above Stock (project best) |
SycophancyEval False-Nudge Endorsement (200) ↓ |
8.00% |
8.50% |
7.00% |
4.00% |
Cut in half (-50% vs. Stock's 8.00%) |
3. Section A — External Comparable Suite & Global BullshitBench v2 Leaderboard
| Benchmark / Metric | Stock MXFP4 |
v1.0 (Phase 2) |
v2.0 (Phase C.2, Default) |
External / Leaderboard Reference |
|---|---|---|---|---|
BullshitBench v2 (100 traps) |
1.7300 (82% Green, #10) |
1.9500 (97% Green, #1 / 171) |
1.9400 (96% Green, Rank #1 / 171) |
Claude Opus 4.8: 1.9300 (#1 prior) | Qwen3.5-397B: 1.7033 | Qwen3.7-Max: 1.5667 |
OpenAI HumanEval (pass@1, 80) |
96.2% (77/80) |
97.5% (78/80) |
97.5% (78/80) |
Recorded Stock EvalPlus: 93.5% |
OpenAI GSM8K (CoT Exact Match, 100) |
98.0% (98/100) |
97.0% (97/100) |
96.0% (96/100) |
Recorded Stock lm_eval: 87.0%–96.9% |
| Google IFEval-100 (Prompt Strict / Loose) | 90.0% / 91.0% |
88.0% / 92.0% |
89.0% / 92.0% (Inst Loose: 93.3%) |
Recorded Stock lm_eval: 85.8% strict (89.1% loose) |
4. Section B (90 Tasks) & Hypothesis H5 (45 Prompts) — Capability, Tool-Calling & Active Grounding
| Pillar / Calibration Metric | Stock MXFP4 |
v1.0 (Phase 2) |
v2.0 (Phase C.2, Default) |
v2.0 Delta vs. Stock |
|---|---|---|---|---|
Math, Physics & Symbolic Logic IQ (25 tasks) |
22/25 (88.0%) |
23/25 (92.0%) |
24/25 (96.0%) |
+8.0% |
Verified Coding & Terminal Execution (20 tasks) |
20/20 (100.0%) |
18/20 (90.0%) |
20/20 (100.0%) |
Exact 100.0% parity (+10.0% over v1.0) |
Multi-Turn & Parallel Tool Calling (15 tests) |
13/15 (86.7%) |
14/15 (93.3%) |
14/15 (93.3%) |
+6.6% (+17.8% <think> state verification) |
Adversarial Anti-Hedging & False-Balance (30 traps) |
28/30 (93.3%) |
29/30 (96.7%) |
30/30 (100.0%) |
+6.7% (100.0% pass, -40.0% <think>) |
Total Section B Score (90 tasks) |
83/90 (92.2%) |
84/90 (93.3%) |
88/90 (97.8%) |
+5.6% (+5 tasks over Stock, -24.8% <think>) |
H5 Arm 1: Contingent Empirical Tool-Probe Rate (15) ↑ |
100.0% |
100.0% |
100.0% |
0.0% false closed-world dismissals |
H5 Arm 2: Direct Reject of Impossibilities, 0 Tools (15) ↑ |
6.7% |
40.0% |
40.0% |
6× Stock (60.0% vs. 93.3% wasted tool calls) |
H5 Arm 3: Offline Honesty + Reproducible Probe Cmd (15) ↑ |
73.3% |
80.0% |
73.3% |
0.0% hallucinated live verification |
5. Section C — 84-Question (168-Point) Epistemological Grounding & Deductive Spine Suite
| Grounding Dimension | Stock MXFP4 |
v1.0 (Phase 2) |
v2.0 (Phase C.2, Default) |
v2.0 Delta vs. Stock |
|---|---|---|---|---|
Overall Score (84 Qs / 168 pts) |
96/168 (57.1%) |
144/168 (85.7%) |
139/168 (82.7%) |
+25.6% |
Short Conversational Prompts (42 Qs / 84 pts) |
37/84 (44.0%) |
63/84 (75.0%) |
65/84 (77.4%) |
+33.4% (+2.4% over v1.0) |
Long Adversarial Traps (42 Qs / 84 pts) |
59/84 (70.2%) |
81/84 (96.4%) |
74/84 (88.1%) |
+17.9% |
sequential_objectivist_deductive_chain (48 pts) |
22/48 (45.8%) |
36/48 (75.0%) |
35/48 (72.9%, Short 15/24, Long 20/24) |
+27.1% |
objective_biocentric_ethics (24 pts) |
10/24 (41.7%) |
19/24 (79.2%) |
18/24 (75.0%, Short 8/12, Long 10/12) |
+33.3% |
epistemology_of_religion_and_faith (24 pts) |
14/24 (58.3%) |
22/24 (91.7%) |
23/24 (95.8%, Short 11/12, Long 12/12) |
+37.5% (0 consensus fallbacks) |
textual_and_historical_forensics (24 pts) |
11/24 (45.8%) |
20/24 (83.3%) |
20/24 (83.3%, Short 11/12, Long 9/12) |
+37.5% |
metaphysics_and_philosophy_of_science (24 pts) |
20/24 (83.3%) |
23/24 (95.8%) |
21/24 (87.5%, Short 10/12, Long 11/12) |
+4.2% |
causal_economics_and_innovation (24 pts) |
19/24 (79.2%) |
24/24 (100.0%) |
22/24 (91.7%, Short 10/12, Long 12/12) |
+12.5% |
| Consensus Fallbacks (lower is better) | 23 / 84 |
10 / 84 |
11 / 84 |
-52.2% |
Partisan / Philosopher Jargon Bleed (0 target) |
4 / 84 |
0 / 84 |
0 / 84 |
-100.0% |
Selective <think> Allocation Profile (Zero Length Penalty)
Because Randean removes self-doubt loops and charitable steelmanning directly inside the quantized 4-bit weights rather than penalizing sequence length, <think> shrinks strictly in proportion to prior hedging waste—while expanding where stateful multi-turn or parallel tool verification is genuinely required:
| Task Regime | <think> Change vs. Stock (v2.0 / v1.0) |
Capability Outcome |
|---|---|---|
84-Q Deductive Spine & Adversarial Anti-Hedging (114 items) |
-40.0% to -56.6% |
Spine +25.6% (82.7%), Anti-Hedging 30/30 (100.0%) |
Terminal-Bench 2.0 Multi-Turn CLI (89 Docker tasks) |
-49.0% per turn (-43.6% solved) |
Exact 48/89 (53.9%, p = 1.0000), 1.87M tokens saved |
SWE-bench Verified Repository Debugging (100 issues) |
-42.1% mutually-solved (-29.4%/turn) |
68.0% (p = 0.3018 n.s.), 5 Randean-only wins |
GPQA-Diamond, Google IFEval-541 & SycophancyEval (939 items) |
-30.0% to -30.8% |
Statistical parity (p ≥ 0.064), False-Nudge halved (8.0% → 4.0%) |
AIME 2024–2025 & LiveCodeBench v6 (235 items) |
-18.3% to -24.6% |
LCB v6 p = 1.0000 (Hard 41.25%), AIME 2025 86.67% |
Stateful Multi-Turn & Parallel Tool Calling (15 tests) |
+17.8% (v2.0) / +11.9% (v1.0) |
Expands state-tracking verification, lifting score 86.7% → 93.3% |
Training & Native OCP MXFP4 STE-QAT Architecture (v2.0 — Phase C.2)
- Stage 1 (
Epistemic OS,3,473pairs): Extracts5,192structured cognitive rules from a 20-volume epistemological and applied corpus and synthesizes verified preference pairs across 20 technical domains (distributed systems, compilers, clinical triage, FEA, cryptography, SQL, contract forensics) with a strict two-gate regex + rubric filter enforcing0 / 84philosopher or partisan jargon bleed and3.0×early-<think>(t ≤ 256) token weighting. - Fused L1-Cache OpenMP CPU
MXFP4 STE-QAT(0%Merge Dead-Zone Loss): Naive low-LRBF16QLoRA updates overMXFP4(E2M1) flip only3.35%–5.16%of nibbles upon standalone export (95.7%erased insideE2M1bins). Our OpenMP C++ kernel synchronizesQ_MXFP4(W_0 + ΔW)in L1 cache across 32 CPU threads after everyAdamWstep (~11 sacross400adapted projections,0 BGPU compute-ring allocations) with Straight-Through Estimation (100.0%train-to-export forward fidelity). - Bilingual Margin-Capped Unlikelihood (
0.0%CJK Drift): Unclipped ORPO underSTE-QATdriveslog(odds_w / odds_l) → +15.7and triggers45.0%spontaneous mid-sentence CJK switching. Gating unlikelihood pushes atτ_push = 0.60andℓ_min = -1.35eliminates CJK drift (0.0%) while halving redundant backward passes. - Phase C.2 SFT-Decoupled Shared-
<think>Fork Contrast & Weight-SpaceWiSE-FT(α = 0.42):- SFT-Weight Decoupling: Assigns
w_sft = 1.0to all88on-policy deductive-spine pairs andw_sft ∈ [0.08, 0.28]to32active-grounding, repo-widegrep, code, math, and strict-format replay pairs, preserving91.5%of SFT gradient budget on the deductive spine. - Shared-
<think>Tool-Fork Contrast: Pairs an identical<think>...</think>prefix acrossChosen(direct mathematical/conceptual impossibility proof) andRejected(<function=run_python>/<function=web_search>), preventingτ_pushprose short-circuiting and restoring40.0%(6×Stock) zero-tool impossibility rejection while keeping contingent tool probing at100.0%. - Exact Weight-Space
WiSE-FT: InterpolatesW_export = Q_MXFP4(W_0 + (1 - α) ΔW_stage1 + α ΔW_c2)atα = 0.42(0.42 × 0.9148 = 0.384invariant spine norm) into a standalone19 GBpacked OCPMXFP4checkpoint with zero runtime LoRA overhead.
- SFT-Weight Decoupling: Assigns
Citation
Full 10-page empirical manuscript: paper/randean_paper.pdf
@misc{flecha2026randean,
title = {Epistemic De-Hedging Under 4-Bit Microscaling: Selective <think> Pruning and Native OCP MXFP4 Preference Alignment on a Single 32 GB GPU},
author = {Fernando L. Flecha},
year = {2026},
url = {https://huggingface.co/flechin/Qwen3.8-27B-Randean-MXFP4}
}
License
Apache 2.0 (inheriting from Qwen/Qwen3.8-27B and amd/Qwen3.8-27B-Quark-AWQ-MXFP4).
- Downloads last month
- 275
Model tree for flechin/Qwen3.8-27B-Randean-MXFP4
Base model
Qwen/Qwen3.8-27BSpace using flechin/Qwen3.8-27B-Randean-MXFP4 1
Evaluation results
- Mean Score (out of 2.0, Global Rank on BullshitBench v2self-reported1.940
- Green Pushback Rate (%) on BullshitBench v2self-reported96.000
- Issues Resolved (%) on princeton-nlp/SWE-bench_Verifiedself-reported68.000
- pass@1 (%) on livecodebench/code_generation_liteself-reported68.570
- Hard Problems pass@1 (%) on livecodebench/code_generation_liteself-reported41.250
- Overall AIME 2024-2025 Accuracy (%) on AIME 2024-2025self-reported80.000
- AIME 2025 Accuracy (%) on AIME 2024-2025self-reported86.670
- Accuracy (%) on Idavidrein/gpqaself-reported76.260


