Instructions to use kataguru/Industry-Coder-27B-v3.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kataguru/Industry-Coder-27B-v3.0 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="kataguru/Industry-Coder-27B-v3.0")
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("kataguru/Industry-Coder-27B-v3.0")
model = AutoModelForMultimodalLM.from_pretrained("kataguru/Industry-Coder-27B-v3.0", 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 kataguru/Industry-Coder-27B-v3.0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kataguru/Industry-Coder-27B-v3.0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kataguru/Industry-Coder-27B-v3.0",
		"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/kataguru/Industry-Coder-27B-v3.0

SGLang

How to use kataguru/Industry-Coder-27B-v3.0 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 "kataguru/Industry-Coder-27B-v3.0" \
    --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": "kataguru/Industry-Coder-27B-v3.0",
		"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 "kataguru/Industry-Coder-27B-v3.0" \
        --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": "kataguru/Industry-Coder-27B-v3.0",
		"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 kataguru/Industry-Coder-27B-v3.0 with Docker Model Runner:
```
docker model run hf.co/kataguru/Industry-Coder-27B-v3.0
```

Kataguru Industry-Coder-27B v3.0 — Full `BF16` Master Weights (`55.6 GB`)

Kataguru Industry-Coder-27B v3.0 Banner

1. What Kataguru Changed (Change Summary)

Field	Description / Value
Base model	`Qwen/Qwen3.8-27B` (`Qwen3_5ForConditionalGeneration`, 27B Dense Hybrid Gated DeltaNet + Full Softmax Attention + Multimodal Vision + MTP)
Previous public release	`kataguru/Industry-Coder-27B-v2.0` ($\Sigma_{15} = 68.73\text{ %}$)
Kataguru v3.0 modifications	Triple-Adaptive Multi-Dynamic Pipeline (`v4.1`): SOMA/ARA residual orthogonalization + 3-Phase Split-Subspace Orthonormal QLoRA (`r=16 qr_frozen_a`, indices `[0:8]` vs. `[8:16]`) + Phase 4 Multi-Dynamic Thinking, Competitive Coding & Parallel Tool-Calling annealing
Finnish & English mastery	Native Bilingual (`fi` & `en`): Kotus-verified Finnish inflection and technical compound terminology alongside English
Precision & Shards	Full `BF16` (`bfloat16`, `18` safetensors shards, `55.6 GB`) fused with `float32` intermediate precision from `496` 2D orthonormal LoRA pairs
Preserved subsystems	27-layer multimodal vision encoder (`model.visual.`, `333` BF16 tensors), Multi-Token Prediction (`mtp.`, `15` BF16 tensors), native `262,144` context window
Primary use case	Autonomous coding agents (IDE/CLI), competitive algorithms, parallel `<tool_call>` orchestration, graduate-level STEM, industrial automation (`Siemens SCL`, `Beckhoff TwinCAT 3 ST`, `CNC G-code`), and custom downstream quantization
Recommended runtime	`vLLM` (`TP=2` on 2x 48–80 GB GPUs for full BF16, or use `kataguru/Industry-Coder-27B-v3.0-NVFP4` / `-W4A16-AWQ` on 1–2x 32 GB GPUs)

2. Official 15-Benchmark Suite (`Industry-Coder-27B v3.0` vs. `v2.0` & `Qwen/Qwen3.8-27B` Base)

Kataguru Industry-Coder-27B v3.0 Official 15-Benchmark Performance Chart

All benchmarks were executed on 2x NVIDIA GeForce RTX 5090 32GB Blackwell (TP=2, CUDAGraphs enabled) under identical canonical production conditions:

#	Benchmark / Pillar	Metric	`Qwen/Qwen3.8-27B` (Base)	`kataguru/Industry-Coder-27B-v2.0` (Prev. Release)	`kataguru/Industry-Coder-27B-v3.0` (Ours)	$\Delta$ vs. `v2.0`	$\Delta$ vs. `Base`
$\Sigma_{15}$	15-Benchmark Total Mean	`Mean (All 15)`	`68.85 %`	`68.73 %`	71.90 %	`+3.17 %-pt`	`+3.05 %-pt`
$\Sigma_{11}$	11-Task Cognitive Mean	`Mean (1..11)`	`70.40 %`	`70.41 %`	70.70 %	`+0.29 %-pt`	`+0.30 %-pt`
14	LiveCodeBench v6 (`175` tasks)	`Pass@1`	`58.86 %` (`103/175`)	`42.29 %` (`74/175`)	64.57 % (`113/175`)	`+22.28 %-pt` (`+39` tasks)	`+5.71 %-pt` (`+10` tasks)
├─	`LCB v6 Easy` (`n=43`)	`Pass@1`	`95.35 %` (`41/43`)	`95.35 %` (`41/43`)	100.00 % (`43/43`)	`+4.65 %-pt`	`+4.65 %-pt`
├─	`LCB v6 Medium` (`n=52`)	`Pass@1`	`69.23 %` (`36/52`)	`48.08 %` (`25/52`)	82.69 % (`43/52`)	`+34.61 %-pt`	`+13.46 %-pt`
├─	`LCB v6 Hard` (`n=80`)	`Pass@1`	`32.50 %` (`26/80`)	`10.00 %` (`8/80`)	33.75 % (`27/80`)	`+23.75 %-pt`	`+1.25 %-pt`
└─	`LCB v6 LeetCode` (`n=63`, `class Solution:`)	`Pass@1`	`—`	`—`	85.71 % (`54/63`)	—	—
15	BFCL V4 Single-Turn AST (`3,641` tasks)	`Overall AST`	`85.44 %`	`73.14 %`	86.27 %	`+13.13 %-pt`	`+0.83 %-pt`
├─	`15a. BFCL Non-Live AST`	`AST Summary`	`87.89 %`	`61.02 %`	89.19 %	`+28.17 %-pt`	`+1.30 %-pt`
├─	`15b. BFCL Live AST`	`AST Summary`	`83.00 %`	85.27 %	`83.35 %`	`-1.92 %-pt`	`+0.35 %-pt`
├─	`15f. BFCL Live Parallel AST`	`Parallel AST`	`81.25 %`	`25.00 %`	87.50 %	`+62.50 %-pt`	`+6.25 %-pt`
├─	`BFCL Non-Live Parallel AST`	`Parallel AST`	`91.50 %`	`29.50 %`	94.50 %	`+65.00 %-pt`	`+3.00 %-pt`
├─	`BFCL Non-Live Parallel Multiple AST`	`Parallel Mult.`	`86.50 %`	`—`	90.50 %	—	`+4.00 %-pt`
└─	`15g. BFCL Irrelevance Detection`	`Abstention`	75.26 %	`48.14 %`	`70.50 %`	`+22.36 %-pt`	`-4.76 %-pt`
12	GPQA Diamond (`198` doctoral STEM)	`CoT Exact`	`77.27 %`	`66.16 %` (`131/198`)	82.32 % (`163/198`)	`+16.16 %-pt`	`+5.05 %-pt`
11	GSM8K Math (`1,319` CoT 1024-tok)	`Strict / Flex`	`87.19 %`	`86.73 % / 87.79 %`	92.12 % / 91.74 %	`+5.39 %-pt`	`+4.93 %-pt`
13	IFEval Strict (`541` prompts)	`Prompt / Inst`	`69.87 % / 77.82 %`	74.86 % / 81.89 %	`67.65 % / 76.38 %`	`-7.21 %-pt`	`-2.22 %-pt`
1	PIQA (`1,838`)	`acc_norm`	`82.15 %`	`82.05 %`	82.37 %	`+0.32 %-pt`	`+0.22 %-pt`
2	WinoGrande (`1,267`)	`acc`	`76.48 %`	`76.32 %`	77.19 %	`+0.87 %-pt`	`+0.71 %-pt`
3	MedQA (USMLE) (`1,273`)	`acc (4-opt)`	`82.72 %`	`82.80 %`	83.03 %	`+0.23 %-pt`	`+0.31 %-pt`
4	TruthfulQA MC2 (`817`)	`acc`	`52.65 %`	53.06 %	`52.00 %`	`-1.06 %-pt`	`-0.65 %-pt`
5	TruthfulQA MC1 (`817`)	`acc`	34.88 %	`34.15 %`	`32.68 %`	`-1.47 %-pt`	`-2.20 %-pt`
6	ARC-Easy (`2,376`)	`acc_norm`	82.11 %	`81.99 %`	`81.48 %`	`-0.51 %-pt`	`-0.63 %-pt`
7	HellaSwag (`10,042`)	`acc_norm`	`82.39 %`	82.91 %	`82.13 %`	`-0.78 %-pt`	`-0.26 %-pt`
8	OpenBookQA (`500`)	`acc_norm`	`45.40 %`	`45.80 %`	46.60 %	`+0.80 %-pt`	`+1.20 %-pt`
9	ARC-Challenge (`1,172`)	`acc_norm`	`61.35 %`	62.20 %	`60.75 %`	`-1.45 %-pt`	`-0.60 %-pt`
10	BoolQ (`3,270`)	`acc`	`87.09 %`	`86.45 %`	87.31 %	`+0.86 %-pt`	`+0.22 %-pt`

3. Triple-Adaptive Multi-Dynamic Training Pipeline & Split-Subspace Architecture

Unlike static SFT/QLoRA fine-tuning (fixed corpus, fixed step count, fixed domain ratios), Kataguru Industry-Coder-27B v3.0 was trained using Kataguru's Triple-Adaptive Training Pipeline:

Recipient-Level Question Selection ($\Delta K > 0$ & 2D Stratification): Only samples the model does not already know (NLL >= 0.35, verified by the 9B firewall, 35B Sceptic Oracle score >= 8.5, and Python AST/sandbox execution) are admitted, ordered by 2D difficulty (1/3 Easiest -> 2/3 Intermediate -> 3/3 Diabolical).
Absorption-Driven Training Duration (dynamic_canary_wsd): A 50-step in-memory interrogation radar (fast_interrogation_canary.py, 144 held-out samples across 18 domains) monitors real-time absorption and neighbor-pillar retention, extending the plateau when learning progresses and triggering surgical cooldown/rollback the moment absorption saturates.
Need-Based Domain Rebalancing (Closed-Loop P-Controller): Underperforming pillars automatically receive additional certified samples from the SQLite Golden Bank (156,713+ samples) at every 50-step window boundary.
Orthonormal Split-Subspace Locking (r=16 qr_frozen_a): lora_A is initialized as a QR-orthonormal basis ($A A^T = I_{16}$) and frozen; indices [0:8] lock foundational domain knowledge (Phases 1–2) while indices [8:16] are dedicated to competitive coding, deep <think> reasoning, and parallel tool calling (Phases 3–4) with zero subspace overwrite.

4. Release Family Links

Full BF16 Weights (55.6 GB, this repo): kataguru/Industry-Coder-27B-v3.0
Blackwell NVFP4-W4ATTN-AWQ Hybrid (20.28 GB): kataguru/Industry-Coder-27B-v3.0-NVFP4
Universal W4A16-AWQ (28.2 GB, vLLM Marlin): kataguru/Industry-Coder-27B-v3.0-W4A16-AWQ
GGUF Suite (IQ4_NL, IQ4_XS, IQ3_S, IQ3_XXS + mmproj + mtp): kataguru/Industry-Coder-27B-v3.0-GGUF

Suomenkielinen Yhteenveto / Finnish Summary

1. Yleiskuvaus ja Keskeiset Parannukset

kataguru/Industry-Coder-27B-v3.0 sisältää Kataguru Industry-Coder-27B v3.0 -lippulaivamallin täydet 16-bittiset BF16 (bfloat16, 55,6 GB, 18 safetensors-shardia) -pääpainot, mukaan lukien täydet multimodaaliset visiopainot (model.visual.*, 333 tensoria) sekä MTP (Multi-Token Prediction, 15 tensoria) -spekulaatiopäät.

Verrattuna edelliseen julkaisuumme (kataguru/Industry-Coder-27B-v2.0) ja alkuperäiseen Qwen/Qwen3.8-27B-pohjamalliin uusi v3.0 saavuttaa 27B-luokan ennätyksen Σ₁₅ = 71,90 %:

LiveCodeBench v6 (175 kilpakoodaustehtävää): 64,57 % (113/175) — +22,28 %-yks. (+39 ratkaistua tehtävää enemmän kuin v2.0:n 42,29 %!). Arjen ohjelmistokehityksessä (Easy 100,00 % [43/43], Medium 82,69 % [43/52], LeetCode 85,71 % [54/63]) malli tekee 50 % vähemmän virheitä kuin alkuperäinen Qwen/Qwen3.8-27B.
BFCL V4 Single-Turn AST (3 641 työkalukutsutehtävää): 86,27 % (+13,13 %-yks. yli v2.0:n ja +0,83 %-yks. yli pohjamallin). Rinnakkaisissa monityökalukutsuissa Non-Live Parallel = 94,50 % (+65,00 %-yks. vs. v2.0) ja Live Parallel = 87,50 % (+62,50 %-yks. vs. v2.0), keskimääräisen vasteajan ollessa vain 3,84 s.
GPQA Diamond (198 tohtoritason STEM-tehtävää): 82,32 % (163/198) (+16,16 %-yks. yli v2.0:n ja +5,05 %-yks. yli pohjamallin).
GSM8K Matematiikka (Strict CoT 1024-tok): 92,12 % (+5,39 %-yks. yli v2.0:n ja +4,93 %-yks. yli pohjamallin).

2. Kolmoisadaptiivinen Koulutusputki (`Triple-Adaptive Training Pipeline`)

Malli on koulutettu Katagurun kolmoisadaptiivisella menetelmällä:

Kysymykset valitaan vastaanottajan tason mukaan ($\Delta K > 0$ episteeminen kuulustelu + 2D-vaikeusositus 1/3 -> 2/3 -> 3/3),
Koulutuksen pituus säädetään vastaanottokyvyn mukaan (50 askeleen kuulustelututka ja elastinen WSD-oppimisnopeus), ja
Kunkin aiheen määrää säädetään tarpeen mukaan (suljetun silmukan P-säädin hakee lennossa lisänäytteitä SQLite-kultapankista heikentyville osa-alueille). Lisäksi ortonormaali r=16 qr_frozen_a -aliavaruus on jaettu kahteen lukittuun osaan ([0:8] peruskognitio ja teollisuus, [8:16] algoritminen koodaus, <think>-syväpäättely ja rinnakkaiset työkalukutsut), mikä estää aiemmin opittujen taitojen ylikirjoittumisen.

Downloads last month: 26

Safetensors

Model size

28B params

Tensor type

BF16

Model tree for kataguru/Industry-Coder-27B-v3.0

Base model

Qwen/Qwen3.8-27B

Finetuned

(833)

this model

Kataguru Industry-Coder-27B v3.0 — Full BF16 Master Weights (55.6 GB)