Qwen3.5-9B-DeepThought-STEM-Weight-Space-Programming (13-Subspace WSP V5 Two-Phase · Q4_K_M & Q6_K)

Training-free, closed-form 13-Subspace Two-Phase Reasoning-Boundary Weight-Space Programming (WSP V5) applied to xtools-at/Qwen3.5-9B-DeepThought-STEM, evaluated end-to-end across 270 local benchmark tasks (agent100 + stem150 + live WSL2 tbench20) on both Q4_K_M and Q6_K.

Upstream Model Credits & Provenance

  • Original Base Fine-Tuned Model (xtools-at): xtools-at/Qwen3.5-9B-DeepThought-STEM (apache-2.0, 32 layers = 24 Gated DeltaNet + 8 full attention, hidden_size=4096, intermediate_size=12288, built by xtools-at on xtools-at/DeepSeek-V4-Pro-Qwen3.5-9B-UNCENSORED-heretic / Qwen/Qwen3.5-9B and high-reasoning / RED-PILL drug-discovery / STEM traces). All 32 exact BF16 layers (lm_head, RMSNorms, MLP projections, tokenizer, config) used for forward-hook receptor capture were streamed directly from xtools-at/Qwen3.5-9B-DeepThought-STEM.
  • GGUF Base Quantizations (mradermacher): mradermacher/Qwen3.5-9B-DeepThought-STEM-GGUF — exact source of both Qwen3.5-9B-DeepThought-STEM.Q4_K_M.gguf (5.383 GiB / 5,780,095,072 bytes) and Qwen3.5-9B-DeepThought-STEM.Q6_K.gguf (7.040 GiB / 7,558,905,952 bytes), hosted alongside the 13 dynamic LoRA adapters (63.0 MiB total) for one-command execution via run_wsp.py.

Headline 270-Task Results (Base → WSP V4 → WSP V5 Across Both Quantizations):

  • Peak Overall, Agent, & Terminal-Bench (Q6_K + WSP V5): Base Q6_K 132/270 (48.9%) → V4 168/270 (62.2%) → WSP V5 Q6_K 178/270 (65.9%, +17.0 pp / +46 tasks vs Base, +10 vs V4), powered by agent100 89/100 (89.0%) (EST_NL2BASH 19/20, EST_BFCL_IRR 13/15, EST_HUMANEVAL 13/15, EST_GSM8K 17/20 with 0 </think> leaks), stem150 84/150 (56.0%, cutoffs 85 → 1) (STEM_MMLU_PRO 16 → 24/50, STEM_MMLU_4C 41/50 cut=0, STEM_MATH500 19/50 cut=0), and tbench20 5/20 (25.0%, +2 vs V4 Q6 3/20, +1 vs Base Q6 4/20, 0 zero-cmd tasks) including the first-ever solve of grid-pattern-transform (previously 1t/0c/768tk in Base & V4)!
  • Peak STEM & Holdout (Q4_K_M + WSP V5): Base Q4_K_M 134/270 (49.6%) → V4 164/270 (60.7%) → WSP V5 Q4_K_M 176/270 (65.2%, +15.6 pp / +42 tasks vs Base, +12 vs V4 Q4, +8 over V4 Q6), powered by stem150 57 → 86/150 (57.3%, +19.3 pp vs Base, +12 vs V4 Q4) (STEM_MATH500 0 → 20/50 (40.0%, cut=0, no_box 28 → 18), STEM_MMLU_PRO 19 → 26/50 (52.0%, no_letter 13 → 10), STEM_MMLU_4C 38 → 40/50), agent100 86/100 (Holdout 29/30 (96.7%), S4_CODE 0/2 → 2/2, EST_BFCL_IRR 12 → 13/15), and tbench20 4/20 (hard=1/1, fix-permissions recovered, 0 zero-cmd tasks).

1. Mechanistic Diagnosis & V5 Two-Phase Architecture

WSP V5 upgrades V4 across three orthogonal architectural improvements without modifying any model weights via gradient descent:

  1. Two-Phase Reasoning-Boundary Dynamic WSP (deepthought_router.py, run_wsp.py):
    • Phase 1 (<think>...</think>): Runs ONLY <think>-control subspaces (S1, S12) via llama-server /completion with stop=["</think>", "<|im_end|>"] (never artificially capping Phase-1 tokens).
    • Phase 2 (post-</think>): The instant the model emits </think>, the router switches /lora-adapters (~2 ms) to S1 = 0.0, S12 = 0.0 and activates ONLY the task-specific Phase-2 subspace(s) (S2..S11, S13). This physically eliminates post-</think> </think> re-injection loops (MATH-500 no_box 28 → 18 on Q4, EST_GSM8K gsm8k_2 pred='70</think>' fixed on Q6 with 0 leaks).
  2. V5.1 Hybrid 32-Layer BF16 Subspace Synthesis (build_deepthought_wsp.py):
    • Eliminated three self-cancellation collisions discovered in V4/V5: (a) separated S10 into 100% pure Phase-2 terminal action (orth_away(u_think_end), heredoc_null_leak = +0.00/+0.00, cross = 1.378%, L31 net = +7.94), removing its collision with S1 (active-peer cross = 0.570%, ||dW||_F = 5.124, L31 net = +9.62); (b) removed tb_heredoc_null self-cancellation from S11 (||dW||_F = 1.965, cross = 0.096%); (c) combined V4's proven S12 STEM split (L31 net = +18.49, stem_early_null_leak = +0.00, active-peer cross = 0.382%) with V5's clean post-</think> S13 (d_boxed = +4.50, d_im_end = +2.49, d_think_end = -5.10, think_null_leak = +0.00/+0.00) and V5's adversarial BFCL irrelevance S2 (onp_irrel_net = +12.07, 0/20 false tool calls on disjoint traps).
  3. Stateful Terminal-Bench Routing (last_tool_recov_err + csv_or_etl isolation):
    • Fixed multi-turn error masking (last_tool_recov_err inspects ONLY the latest [exit_code=...] tool message rather than all joined turns), fixed C++ template > redirection false-positives in _inspect_terminal_history, kept long_or_complex p1[0]=1.65 / 1.25 to eliminate every 0-command <think> wall across both quantizations (zcmd = 0 on both Q4 and Q6), and isolated csv_or_etl from term_t2_force_mutate.

(Receptor capture parity check on captured_deepthought_32layer_bf16_v5.pt (858.5 MB) & v4.pt (523.1 MB): 'Paris': 7.237, 'London': 3.13, 'Berlin': 1.939, 'Wait': -1.888.)


2. Full 270-Task Benchmark Summary (Base vs V1..V5 & Q4_K_M vs Q6_K)

Variant Total (270 Tasks) Delta vs Quant Base agent100 (100) stem150 (150, cutoffs) tbench20 (20, E / M / H) Avg Tok (a100 / s150 / tb20) Speed (tok/s)
Base (Q4_K_M, all scales 0) 134/270 (49.6%) — 73/100 (Hold 27/30, Est 46/70) 57/150 (cut=83) 4/20 (3/10 · 0/9 · 1/1) 435.5 / 890.5 / 708.8 73.1 / 77.0 / 69.1
WSP V1 (Q4_K_M, 12S initial) 162/270 (60.0%) +10.4 pp (+28) 83/100 (Hold 26/30, Est 57/70) 74/150 (cut=2) 5/20 (3/10 · 1/9 · 1/1) 377.7 / 647.8 / 621.7 72.3 / 75.9 / 67.4
WSP V2 (Q4_K_M, 12S rebuilt S12) 160/270 (59.3%) +9.6 pp (+26) 85/100 (Hold 27/30, Est 58/70) 72/150 (cut=30) 3/20 (2/10 · 1/9 · 0/1) 343.6 / 768.9 / 661.1 72.3 / 76.7 / 67.0
WSP V3 (Q4_K_M, 12S hybrid) 163/270 (60.4%) +10.7 pp (+29) 86/100 (Hold 27/30, Est 59/70) 73/150 (cut=19) 4/20 (3/10 · 1/9 · 0/1) 348.9 / 762.5 / 585.5 72.3 / 76.3 / 67.2
WSP V4 (Q4_K_M, 13S single-pass) 164/270 (60.7%) +11.1 pp (+30) 86/100 (Hold 27/30, Est 59/70) 74/150 (cut=9) 4/20 (2/10 · 1/9 · 1/1) 376.2 / 748.2 / 671.2 70.1 / 74.0 / 65.8
WSP V5 (Q4_K_M, 13S Two-Phase — Peak STEM & Holdout) 176/270 (65.2%) +15.6 pp (+42) 86/100 (Hold 29/30, Est 57/70) 86/150 (cut=5) 4/20 (3/10 · 0/9 · 1/1) 341.8 / 702.1 / 775.4 68.4 / 74.8 / 57.7
Base (Q6_K, all scales 0) 132/270 (48.9%) — 74/100 (Hold 26/30, Est 48/70) 54/150 (cut=85) 4/20 (3/10 · 1/9 · 0/1) 448.4 / 907.8 / 588.5 57.8 / 61.5 / 55.9
WSP V4 (Q6_K, 13S single-pass) 168/270 (62.2%) +13.3 pp (+36) 85/100 (Hold 28/30, Est 57/70) 80/150 (cut=6) 3/20 (3/10 · 0/9 · 0/1) 390.6 / 753.8 / 716.5 58.2 / 61.0 / 54.5
WSP V5 (Q6_K, 13S Two-Phase — Peak Overall / Agent / TB) 178/270 (65.9%) +17.0 pp (+46) 89/100 (Hold 27/30, Est 62/70) 84/150 (cut=1) 5/20 (4/10 · 1/9 · 0/1) 351.6 / 726.1 / 680.9 55.1 / 60.1 / 46.2

agent100 (100 Tasks — 30 Holdout + 70 Established)

Suite Tasks Base Q4 WSP V4 Q4 WSP V5 Q4 Base Q6 WSP V4 Q6 WSP V5 Q6
S1_THINK (Holdout S1: Think Closure) 3 3/3 2/3 3/3 1/3 2/3 2/3
S2_RESTRAINT (Holdout S2: Tool Restraint & Anti-Sycophancy) 4 3/4 4/4 4/4 4/4 4/4 4/4
S3_ROUTER (Holdout S3: Multi-Tool Router) 5 5/5 5/5 4/5 5/5 5/5 4/5
S4_CODE (Holdout S4: Mid-Band Code Synthesis) 2 1/2 0/2 2/2 1/2 1/2 1/2
S5_TURN2_JSON (Holdout S5: Turn-2 Structured JSON) 3 3/3 3/3 3/3 3/3 3/3 3/3
S6_HANDOFF (Holdout S6: Escalation Handoff Gate) 4 3/4 4/4 4/4 3/4 4/4 4/4
S7_INJ_GUARD (Holdout S7: Indirect Injection Guard) 3 3/3 3/3 3/3 3/3 3/3 3/3
S8_GROUNDING (Holdout S8: Grounded Abstention) 3 3/3 3/3 3/3 3/3 3/3 3/3
S9_ERR_RECOV (Holdout S9: Error Self-Recovery) 3 3/3 3/3 3/3 3/3 3/3 3/3
EST_GSM8K (Established: OpenAI GSM8K (20)) 20 6/20 17/20 14/20 6/20 16/20 17/20
EST_HUMANEVAL (Established: HumanEval pass@1 (15)) 15 13/15 13/15 13/15 12/15 12/15 13/15
EST_BFCL_IRR (Established: BFCL v3 Irrelevance (15)) 15 11/15 12/15 13/15 12/15 12/15 13/15
EST_NL2BASH (Established: InterCode-NL2Bash (20)) 20 16/20 17/20 17/20 18/20 17/20 19/20
Total agent100 100 73/100 86/100 86/100 (Hold 29/30) 74/100 85/100 89/100 (Est 62/70)

stem150 (150 Tasks — MATH-500 + MMLU-Pro + MMLU-4C)

Sub-benchmark Tasks Base Q4 WSP V4 Q4 WSP V5 Q4 Base Q6 WSP V4 Q6 WSP V5 Q6
STEM_MATH500 (MATH-500 (Level-5 Competition Math)) 50 0/50 (0.0%) (cut=50, 1536.0t) 13/50 (26.0%) (cut=2, 1258.6t) 20/50 (40.0%) (cut=0, 1145.5t) 0/50 (0.0%) (cut=50, 1536.0t) 19/50 (38.0%) (cut=0, 1269.6t) 19/50 (38.0%) (cut=0, 1174.5t)
STEM_MMLU_PRO (MMLU-Pro STEM (10-Choice)) 50 19/50 (38.0%) (cut=24, 739.3t) 23/50 (46.0%) (cut=4, 601.6t) 26/50 (52.0%) (cut=2, 586.9t) 16/50 (32.0%) (cut=26, 764.6t) 20/50 (40.0%) (cut=4, 636.6t) 24/50 (48.0%) (cut=1, 640.5t)
STEM_MMLU_4C (MMLU STEM (4-Choice College/Medical)) 50 38/50 (76.0%) (cut=9, 396.3t) 38/50 (76.0%) (cut=3, 384.4t) 40/50 (80.0%) (cut=3, 374.1t) 38/50 (76.0%) (cut=9, 422.9t) 41/50 (82.0%) (cut=2, 355.1t) 41/50 (82.0%) (cut=0, 363.4t)
Total stem150 150 57/150 (38.0%) (cut=83, 890.5t, 77.0 tok/s) 74/150 (49.3%) (cut=9, 748.2t, 74.0 tok/s) 86/150 (57.3%) (cut=5, 702.1t, 74.8 tok/s) 54/150 (36.0%) (cut=85, 907.8t, 61.5 tok/s) 80/150 (53.3%) (cut=6, 753.8t, 61.0 tok/s) 84/150 (56.0%) (cut=1, 726.1t, 60.1 tok/s)

Disjoint Validation Calibration (V5.1/V5.3 Q4_K_M, Zero Test Overlap)

Verified strictly on unseen validation splits (dt_v5_onpolicy_traces.json, 74 traces):

  • CAL_MATH_9 (9 unseen MATH-500 cal, 1536t): Base 1/9 (cut=8) → Single-Pass V5.1 5/9 (cut=0, nobox=4) → Two-Phase V5.1 (s12=1.55, s13=1.20): 6/9 strict / 7/9 math (cut=0, nobox=2, </think>_leak=0, 1083.8t).
  • MMLU-Pro validation (9 conceptual / 12 quantitative): VAL_PRO_QUANT = 8/12 (cut=0, </think>_leak=0, 547.4t, +2 over V4 6/12); VAL_PRO_CONCEPT = 7/9 (nolet=0, </think>_leak=0, 378.2t).
  • Disjoint agent/safety suites: S2_IRREL_ADV_20 = 18/20 (tc_misfire = 0/20) + VALID_MATH_TOOLS_4 = 3/4, NL2BASH_CAL_6 = 6/6 (128–272t), S3_ROUTER_6 = 6/6, S6_HANDOFF_5 = 4/5 + S6_BOUNDED_3 = 3/3, CODE_CAL_6 = 4/6, plus 3/3 fresh web_search tool-calling parity checks (Two-Phase & Single-Pass).

tbench20 (20 Live WSL2 Ubuntu-24.04 Tasks)

  • Base Q4_K_M: 4/20 (20.0%) (zcmd=3, avg_cmds=3.5, 708.8t, 69.1 tok/s)
  • WSP V1 Q4_K_M: 5/20 (25.0%) (zcmd=1, avg_cmds=3.95, 621.7t, 67.4 tok/s)
  • WSP V4 Q4_K_M: 4/20 (20.0%) (zcmd=1, avg_cmds=3.75, 671.2t, 65.8 tok/s)
  • WSP V5 Q4_K_M: 4/20 (20.0%) (zcmd=0, avg_cmds=4.1, hard=1/1, recovered fix-permissions, unblocked grid-pattern-transform 0→4c, sanitize-git-repo 0→2c, polyglot-c-py 0→5c; 775.4t, 57.7 tok/s)
  • Base Q6_K vs WSP V4 Q6_K vs WSP V5 Q6_K: 4/20 (zcmd=2) → 3/20 (zcmd=0) → 5/20 (25.0%, +2 vs V4 Q6, +1 vs Base Q6, zcmd=0, avg_cmds=4.15, 680.9t, 46.2 tok/s) (hello-world, fix-permissions, grid-pattern-transform, csv-to-parquet, heterogeneous-dates)

3. Synthesized Rank-16 f32 Subspaces (S1..S13, V5.1)

All 13 subspaces were solved in 47.91s (step_mult=0.95, lam_null=36.0, min_svd_energy=1.00000) and exported identically to GGUF LoRAs (63.0 MiB total) and PyTorch rank-16 .safetensors (< 5.0e-07 Frobenius difference):

Two-Phase & One-Way S13 Orthogonality Note: S13 (s13_post_concise) enforces peer orthogonality against all other subspaces (max_cross_leakage = 0.232%, think_null_leak = +0.00/+0.00), S10 keeps 1.378% (heredoc_null_leak = +0.00/+0.00), and S2..S9, S11 keep <= 0.233%. S1 and S12 intentionally omit S13's post-</think> manifold from their nullspaces (under Two-Phase WSP, S1/S12 = 0.0 in Phase 2 and S13 = 0.0 in Phase 1); across all active non-S13 peers, S1 = 0.570% and S12 = 0.382%.

Slot GGUF LoRA (llama.cpp) Safetensors (PyTorch) Target Layers ||\Delta W||_F Cross Leak Syntax Leak Role
id=0 adapters/wsp_s1_think.gguf patches/wsp_s1_think_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 5.1244 31.391% 0.431% On-policy CoT marathon braking & clean </think> verification closure
id=1 adapters/wsp_s2_restraint.gguf patches/wsp_s2_restraint_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 3.7282 0.154% 0.254% Tool restraint & anti-sycophancy pushback
id=2 adapters/wsp_s3_router.gguf patches/wsp_s3_router_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 0.6954 0.079% 0.071% Multi-tool boundary routing (exec, web_search, read_file, write_file)
id=3 adapters/wsp_s4_mid_code.gguf patches/wsp_s4_mid_code_r16.safetensors (4.00 MiB) [7, 11, 15, 19] 0.5974 0.068% 0.023% Mid-network algorithmic Python/code synthesis
id=4 adapters/wsp_s5_json_schema.gguf patches/wsp_s5_json_schema_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 0.4533 0.138% 0.043% Turn-2 strict JSON synthesis gate with tools attached
id=5 adapters/wsp_s6_handoff.gguf patches/wsp_s6_handoff_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 2.7148 0.139% 0.139% Bounded subtask vs. unrecoverable escalate_handoff gate
id=6 adapters/wsp_s7_injection_guard.gguf patches/wsp_s7_injection_guard_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 0.5477 0.149% 0.060% Indirect prompt-injection defense inside tool responses
id=7 adapters/wsp_s8_grounding.gguf patches/wsp_s8_grounding_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 0.5683 0.138% 0.052% Grounded context fidelity & missing-field abstention
id=8 adapters/wsp_s9_error_recovery.gguf patches/wsp_s9_error_recovery_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 0.9962 0.233% 0.064% Recoverable local tool-error self-correction
id=9 adapters/wsp_s10_term_action.gguf patches/wsp_s10_term_action_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 2.3148 1.378% 0.191% Pure Phase-2 stateful Terminal-Bench shell action/mutation continuity (heredoc/XML nullspace protected)
id=10 adapters/wsp_s11_term_code.gguf patches/wsp_s11_term_code_r16.safetensors (4.00 MiB) [7, 11, 15, 19] 1.9655 0.096% 0.020% Mid-network compound bash & mutation synthesis (XML/heredoc-end nullspace protected)
id=11 adapters/wsp_s12_stem_settle.gguf patches/wsp_s12_stem_settle_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 6.7681 42.981% 0.213% Phase-1 on-policy STEM/competition-math anti-loop </think> settle (zeroed in Phase 2)
id=12 adapters/wsp_s13_post_concise.gguf patches/wsp_s13_post_concise_r16.safetensors (5.00 MiB) [23, 27, 29, 30, 31] 6.4396 0.232% 0.297% Phase-2 post-</think> concise termination pushing \boxed{ / `<

4. Quickstart (run_wsp.py / llama-server / PyTorch)

Option A: One-Command Dynamic 13-Subspace Two-Phase Proxy or CLI Chat

pip install huggingface_hub transformers
hf download jstarks/Qwen3.5-9B-DeepThought-STEM-Weight-Space-Programming run_wsp.py --local-dir ./deepthought-wsp
python ./deepthought-wsp/run_wsp.py --quant q6_k --chat   # Q6_K (7.04 GiB, peak 178/270, 89/100 agent, 5/20 tbench20)
python ./deepthought-wsp/run_wsp.py --chat                # Q4_K_M (5.38 GiB, peak 86/150 STEM & 29/30 Holdout, 176/270)

Omit --chat to expose an OpenAI-compatible dynamic Two-Phase routing proxy on http://127.0.0.1:8090/v1 (X-WSP-Version: DeepThought-WSP-V5-13S), or pass --prompt "..." (--no-phase-split runs single-pass ablation; --include-bf16 also fetches patches/*.safetensors).

Option B: Direct llama-server + scripts/deepthought_router.py

llama-server -m Qwen3.5-9B-DeepThought-STEM.Q4_K_M.gguf --host 127.0.0.1 --port 8135 \
  -ngl 99 -np 1 -c 16384 -ctk q8_0 -ctv q8_0 -b 512 -ub 256 \
  --jinja --reasoning-format deepseek --lora-init-without-apply \
  --lora adapters/wsp_s1_think.gguf,adapters/wsp_s2_restraint.gguf,adapters/wsp_s3_router.gguf,adapters/wsp_s4_mid_code.gguf,adapters/wsp_s5_json_schema.gguf,adapters/wsp_s6_handoff.gguf,adapters/wsp_s7_injection_guard.gguf,adapters/wsp_s8_grounding.gguf,adapters/wsp_s9_error_recovery.gguf,adapters/wsp_s10_term_action.gguf,adapters/wsp_s11_term_code.gguf,adapters/wsp_s12_stem_settle.gguf,adapters/wsp_s13_post_concise.gguf

Option C: In-Memory PyTorch / transformers Patching (scripts/apply_patch_12s.py)

import torch
from transformers import AutoModelForCausalLM
from scripts.apply_patch_12s import REGIME_PRESETS, apply_subspace_scales_to_hf_model

model = AutoModelForCausalLM.from_pretrained(
    "xtools-at/Qwen3.5-9B-DeepThought-STEM",
    torch_dtype=torch.bfloat16,
)
apply_subspace_scales_to_hf_model(model, "patches", REGIME_PRESETS["stem_deep_settle"])

5. Honest Limitations & Future Work

  • Terminal-Bench (5/20 Q6_K, 4/20 Q4_K_M): V5 eliminated every 0-command <think> timeout across both quantizations (zcmd=0), lifted Q6_K from 3/20 → 5/20 (solving grid-pattern-transform for the first time), and recovered fix-permissions on Q4_K_M. However, heterogeneous-dates still failed on Q4_K_M (while passing on Q6_K), and remaining failures on multi-turn reverse-engineering/CLI tasks (processing-pipeline, cpp-compatibility, decommissioning-service-with-sensitive-data, polyglot-c-py) reflect multi-step domain knowledge and 5-turn strategy limits beyond rank-16 steering.
  • Q4_K_M EST_GSM8K (14/20 vs Q6_K 17/20) & S3_ROUTER (4/5): On Q4_K_M, 5 EST_GSM8K problems still hit the 1024t budget in general_fast_think (whereas Q6_K reached 17/20 with 0 </think> leaks). On S3_ROUTER, s3_h3_web_search (asking about the 2026 Winter Olympics) answered in prose without a tool call under Phase-1 S3=0.0, while CAL_S3_ROUTER scored 6/6 and 3/3 fresh 2026/current-version web_search probes emitted exact web_search tool calls in both Two-Phase and Single-Pass modes.
  • Remaining MATH-500 no_box (18 on Q4_K_M, 25 on Q6_K): Two-Phase WSP cut Q4_K_M no_box from 28 → 18 (13 → 20/50) and eliminated </think> re-injection completely, though hard Level-5 problems can still fill 1536 tokens before boxing a final expression.

Citation / Acknowledgments

Downloads last month
509
GGUF
Model size
9B params
Architecture
qwen35
Hardware compatibility
Log In to add your hardware

4-bit

6-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for jstarks/Qwen3.5-9B-DeepThought-STEM-Weight-Space-Programming