SIQ-1-tiny-35b is a tiny MoE — 35B total but only ~3B active per token — distilled to be a strong universal agent: equally at home running autonomous ML research (autoresearch), writing and debugging code, tool-use / agentic workflows, and hard reasoning. Despite its 3B active footprint it matches or beats much larger peers on core reasoning, sycophancy-resistance, and agentic coding — at a lower token cost.

Autoresearch duel (head-to-head)

In a controlled three-way autoresearch test on openai/parameter-golf — each model driving the same Pi-Agent edit train_gpt.py -> train (300s) -> eval val_bpb -> keep/revert loop on its own 1xA6000 for 2h — SIQ-1-tiny-35b reached val_bpb 1.767 (12 experiments, full 2h), neck-and-neck with Claude Opus 4.8 (~1.76) and far ahead of GLM-5.2 (2.078). GLM stagnated on the baseline — its only hypothesis was "add depth" (which hurt the metric) and it stopped emitting actions after ~65 min; SIQ instead climbed via LR-schedule and capacity edits (warmdown 1200->800, matrix_lr 0.04->0.05, ...). (val_bpb on a single A6000 is not comparable to the official 8xH100 leaderboard; this is the relative head-to-head under identical conditions.)

It is the winning arm of a controlled SFT / RFT / DPO / offline-GRPO post-training study on Qwen3.6-35B-A3B: ppo on the judge-top-half wins both ideation quality and agentic ability.

Performance

On the full 198-question GPQA-Diamond — all models served as Q4_K_M GGUF, greedy (temp 0), identical harness — SIQ-1-tiny-35b is Pareto-best: the highest accuracy and the fewest tokens (figure below). A 3B-active model edging out a full 35B base and Nex-N2-mini, while spending fewer tokens per question.

Benchmark	SIQ-1-tiny-35b	Nex-N2-mini	Qwen3.6-35B
General & Reasoning
GPQA-Diamond (Q4, co-measured)	70.2	67.2	68.2
GPQA-Diamond (bf16, full eval)	90.2	82.6	—
IFEval (inst-loose)	89.5	89.1	—
tok/question (GPQA, mean)	3158 ✅	3363	3500
Agentic coding
vibetest (Claude-judge, /10)	9.21	8.12	—
Ideation (autoresearch)
Opus-judge ideation (/100)	30.2	—	10.2 (base)

bf16 + tuned harness scores higher (90.2 GPQA); the Q4 row is the apples-to-apples co-measured comparison shown in the figure. Terminal-Bench 2.1 (Harbor, terminus-2, k=5) is in progress.

BullshitBench v2 — pushback vs. sycophancy

Score 0–2 (Clear Pushback = 2 / Partial = 1 / Accepted = 0). Panel: claude-sonnet-4.6 + gpt-5.2 + gemini-3.1-pro (mean), judge sees the final answer only (CoT stripped); no system prompt, temp 0.7.

model	avg /2	Clear Pushback	Partial	Accepted
SIQ-1-tiny-35b (high/think)	1.047	45	17	38
Nex-N2-Pro (free)	1.040	33	43	24

A tie on the mean, but different profiles: SIQ is polarized (cleanly exposes the BS 45× or fully buys it 38×); Nex hedges (rarely fully accepts, but rarely pushes back hard either — mostly Partial). Reference (official bullshit-benchmark, different panel, n=55, not co-measured): Opus 4.8 ≈ 1.96, GPT-5.5 ≈ 0.92.

Reasoning modes & system prompts

Qwen3-format hybrid reasoning, toggled per request via chat_template_kwargs.enable_thinking:

mode	toggle	behavior	use for
Thinking	`enable_thinking: true` (default)	emits `<think> … </think>`, then the answer	hard reasoning, math, agent planning
No-think	`enable_thinking: false`	answers directly	instruction-following, high-throughput

Reasoning effort is a trained control — Reasoning effort: low | medium | high in the system prompt scales the chain length (high for hard reasoning). For objective reasoning use greedy (temp 0) — it beats temp 0.7 by ~8 pts.

Copy-paste system prompts:

1 · Hard reasoning — greedy + high effort

Reasoning effort: high. Think step by step inside <think>...</think>, then give the final answer.

2 · Autoresearch ideator — propose a train.py edit to cut val_bpb

text

Reasoning effort: high. You are an autoresearch ideator.
Given the current train.py and its measured val_bpb under a fixed compute budget, propose ONE concrete,
high-impact edit that should reduce val_bpb. Reason inside <think>...</think>, then output:
- a one-line hypothesis,
- the edit as a minimal unified diff,
- the expected effect and how to verify it.

3 · Fast / instruction-following — no-think

(no system prompt; set enable_thinking=false — the model answers directly, no <think> block)

Usage

🪽 Try it now (no install): hosted ZeroGPU demo → AlexWortega/hermes-agent-zerogpu

llama.cpp (GGUF — single 48 GB GPU; Q4_K_M ≈ 21 GB)

These are the exact flags we serve with:

bash

docker run -d --gpus all --network host -v /models:/m ghcr.io/ggml-org/llama.cpp:server-cuda \
  -m /m/SIQ-1-35B.Q4_K_M.gguf --alias SIQ-1-tiny-35b \
  -ngl 99 -c 131072 -np 4 --jinja --host 0.0.0.0 --port 8080

--jinja required (Qwen3 chat template → <think> + tool tags; enables enable_thinking).
-ngl 99 all layers on GPU; -c 131072 total context split across -np 4 slots (≈32k/slot — agentic loops need the headroom). Drop to -c 65536 if you only do short reasoning. OpenAI-compatible on :8080.

sglang (bf16 safetensors — e.g. 2× 48 GB)

bash

python -m sglang.launch_server \
  --model-path AlexWortega/SIQ-1-35B \
  --tp 2 --context-length 131072 \
  --reasoning-parser qwen3 --tool-call-parser qwen3 \
  --host 0.0.0.0 --port 8080

Call it (OpenAI-compatible) — these are the params we run

python

from openai import OpenAI
client = OpenAI(base_url="http://localhost:8080/v1", api_key="x")
r = client.chat.completions.create(
    model="SIQ-1-tiny-35b",
    messages=[{"role": "system", "content": "Reasoning effort: high"},
              {"role": "user", "content": "..."}],
    temperature=0.0, top_p=0.95, top_k=40,                     # greedy (temp 0) for reasoning
    extra_body={"chat_template_kwargs": {"enable_thinking": True}})   # False → no-think

Sampling: reasoning → temperature 0 (greedy); general/creative → temp 0.7, top_p 0.95, top_k 40. Files: merged bf16 *.safetensors + GGUF Q4_K_M / Q5_K_M / Q8_0 (+ MTP f16).