---
title: KnowLine-35B-A3B-Gen1.5
canonical_url: "https://www.modelscope.ai/models/EMLab/KnowLine-35B-A3B-Gen1.5"
md_url: "https://www.modelscope.ai/models/EMLab/KnowLine-35B-A3B-Gen1.5.md"
repository: EMLab/KnowLine-35B-A3B-Gen1.5
last_updated: 2026-11-13
license: apache-2.0
pipeline_tag: text-classification
tasks:
  - text-classification
model_type:
  - qwen3_5_moe
architectures:
  - Qwen3_5MoeForConditionalGeneration
base_model:
  - Qwen/Qwen3.5-35B-A3B
base_model_relation: finetune
parameters: 36.0B
tensor_type:
  - BF16
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - Pytorch
language:
  - en
  - zh
downloads: 23
stars: 0
tags:
  - decision-model
  - system-one
  - decision-index
  - lora-merged
  - moe
---

# KnowLine-35B-A3B-Gen1.5

> KnowLine-35B-A3B-Gen1.5 - An open-source model by EMLab on ModelScope. KnowLine-35B-A3B-Gen1.5

EMLab/KnowLine-35B-A3B-Gen1.5 is a 36.0B-parameter text-classification model on ModelScope. licensed under apache-2.0. derived from Qwen/Qwen3.5-35B-A3B.

- **Repository**: EMLab/KnowLine-35B-A3B-Gen1.5
- **License**: apache-2.0
- **Tasks**: text-classification
- **Parameters**: 36.0B
- **Base model**: Qwen/Qwen3.5-35B-A3B
- **Tags**: decision-model, system-one, decision-index, lora-merged, moe
- **Downloads**: 23
- **Stars**: 0
- **Last updated**: 2026-11-13

Source: https://www.modelscope.ai/models/EMLab/KnowLine-35B-A3B-Gen1.5

---

# KnowLine-35B-A3B-Gen1.5

**KnowLine-35B-A3B-Gen1 with a new way of serving it: the same weights, and a front end that reads short requests twice
before it answers.**

[Inference guide](INFERENCE.md) · [中文说明](README.zh.md) · weights Apache-2.0 · previous version:
[KnowLine-35B-A3B-Gen1](https://huggingface.co/PelaAI/KnowLine-35B-A3B-Gen1)

We raise the minor version when only inference changes and the major version (Gen) when the weights change. Every weight,
config and tokenizer file of Gen1.5 is byte-identical to Gen1.

Highlights:

- **Same weights, read twice.** For a text-only request of up to 3,000 tokens, the front end writes the whole input a
  second time before the model answers. Nothing was trained; see [What changed in Gen1.5](#what-changed-in-gen15).
- **Decision Index 0.3 public suite: 70.72** (self-run, fresh full run), up from 69.49 for Gen1 with the same weights.
  The gains are concentrated in reasoning and inference over long inputs (knowledge and reasoning 52.5 → 55.0, language
  71.2 → 72.9); short-text tasks are flat or slightly lower. See [Comparison](#comparison).
- **Private scores:** the board's full score comes from the maintainers' private run. Gen1 is listed at 61.20. Gen1.5 has
  the same weights as Gen1; its private scores will be produced by the maintainers' run.
- **Other public benchmarks:** C-Eval 88.1 (Gen1 on the same files: 87.7), JevBench / JevBench-hard 87.0 / 73.0
  (85.7 / 70.3); see [Evaluation](#evaluation).
- **Calibration:** the temperature folded into the weights, fitted on held-out data, is unchanged (T = 1.20). Computed with the board's method,
  ECE is about 0.03-0.04 (Gen1: 0.04-0.05) and the Brier score about 0.24-0.26 (Gen1: 0.25-0.27); see
  [Calibration](#calibration).
- **Cost:** about 1.56 times the prompt tokens of Gen1 on the Decision Index suite. One request at a time on one H20:
  63.6 ms median and 101 ms mean, against 51 ms and 73 ms for Gen1 (our measurement, not the board's reference
  hardware); see [INFERENCE.md](INFERENCE.md).
- **Images: unchanged.** Requests with images are sent exactly as in Gen1, so Gen1's image results apply; see
  [Images](#images-zero-shot).

KnowLine is an independent model. It is not affiliated with, endorsed by, or derived from TypeSafe or Jev.

## What it does

You send a state (text or a chat) and up to 64 typed questions: yes/no, choose one of k, or score on a rubric. The model
answers every question with a probability distribution over its options:

- one forward pass per question, with no text generation;
- the output is the probability of each option's label token;
- existing Jev clients only need to change the base URL, and users of KnowLine-35B-A3B-Gen1 only need to change the
  model name, because the interface is the same.

| | |
|---|---|
| Base model | [Qwen/Qwen3.5-35B-A3B](https://huggingface.co/Qwen/Qwen3.5-35B-A3B) (Apache-2.0): mixture of experts, 35B parameters in total, about 3B active per token |
| Weights | byte-identical to [KnowLine-35B-A3B-Gen1](https://huggingface.co/PelaAI/KnowLine-35B-A3B-Gen1): LoRA SFT (rank 32, alpha 64, language model only, including the routed experts), merged into bf16 weights (about 67 GB), with a readout temperature T = 1.20 folded into the final norm weight |
| Serving | `serve_knowline.sh`: SGLang converts the weights to FP8 at load (about 35 GB); `knowline_server.py` reads short requests twice (on by default; `--no-read-twice` gives the Gen1 prompts). We ran every evaluation with one server on one 96 GB H20; plan for at least 48 GB of GPU memory. |
| Languages | English, Simplified Chinese, Traditional Chinese |

## Quickstart

```bash
pip install "sglang==0.5.21" "transformers==5.12.1" requests
bash serve_knowline.sh PelaAI/KnowLine-35B-A3B-Gen1.5 0 8080   # SGLang (FP8 at load) on :9080 + /v1/systemone on :8080
curl -s http://127.0.0.1:8080/v1/systemone -H 'Content-Type: application/json' -d '{
  "model": "m",
  "state": "Customer: my order arrived broken, I want my money back.",
  "questions": {"refund": {"type": "noul", "instructions": "Should the agent offer a refund?"},
                "tone": {"type": "choice", "instructions": "Customer tone?",
                         "criteria": {"angry": "Angry", "neutral": "Neutral", "happy": "Happy"}}}}'
```

- **Front end:** `knowline_server.py` with `knowline_layout.py` next to it; together they depend only on transformers and
  requests.
- **Full settings:** the exact settings used for our Decision Index run are in [INFERENCE.md](INFERENCE.md).

## What changed in Gen1.5

1. **Read twice.** Gen1 saw each request once: the state, then the instruction, the question and its options. Gen1.5
   writes all of that a second time, introduced by the line "Read the same input again before answering:", and then
   reads the answer from the label token as before. The rule is mechanical: it applies to every text-only request of at
   most 3,000 tokens (94.6% of the Decision Index requests); longer requests and requests with images are sent as in
   Gen1.
2. **A likely reason.** With causal attention, the first copy of a long state is read before the model knows the
   question; in the second copy every token can also attend to the first copy, question included. This fits where the
   gains are: reasoning and inference over long inputs.
3. **No training.** The weights, the temperature and the label-token readout are those of Gen1. We checked the change on
   a fresh full run of the Decision Index public suite before release.
4. **Credit.** The read-twice layout follows FlexDM ([github.com/1000luca/FlexDM](https://github.com/1000luca/FlexDM),
   MIT); repeating the input goes back to RE2 (Xu et al., 2023,
   [arXiv:2309.06275](https://arxiv.org/abs/2309.06275)). `knowline_layout.py` is our own implementation.

## Comparison

Scores on the Decision Index. Since Decision Index 0.4, the board's full score is 60% private tests of the same skills
and 40% private tasks from new domains; the public suite is shown but no longer counts. Gen1.5 has no board entry yet.

| model | size | DI 0.3 public | DI 0.4.1 full | source |
|---|---|---|---|---|
| **KnowLine-35B-A3B-Gen1.5** | 35B-A3B | **70.72** | not yet scored | self-run, official kit |
| KnowLine-35B-A3B-Gen1 (same weights) | 35B-A3B | 69.49 | 61.20 | board |
| Fastino GLiDE V1.2 | – | 71.99 | 71.52 | board |
| Jeffan-31b-1710 | 31B | 73.13 | 69.75 | board |
| mammoth-1 | 27B | 75.18 | 68.87 | board |
| Perplexity Decider v1.2 | 27B | 67.74 | 67.98 | board |
| StartLux-Decision v1.1 27B | 27B | 72.88 | 67.92 | board |
| FlexDM · Gemma-4-31B | 31B | 63.63 | 67.56 | board |
| KnowLine-26B-A4B-Gen2 | 26B-A4B | 66.94 | 61.34 | board |
| Jev 1.14 | (API) | 60.70 | 60.24 | board |
| JEV-35B (Qwen3.5-35B-A3B) | 35B-A3B | 59.73 | 57.61 | board |
| Tern-35B-A3B | 35B-A3B | 59.01 | 56.62 | board |

Board rows from the board's data of 2026-10-21 (22:34 UTC, Decision Index 0.4.1).

Public-suite skill by area (DI 0.3, same weights):

| area | Gen1.5 | Gen1 |
|---|---|---|
| Knowledge & reasoning | 55.0 | 52.5 |
| Language | 72.9 | 71.2 |
| Retrieval & routing | 75.4 | 75.2 |
| Tools & agents | 88.2 | 88.1 |
| Arts & taste | 64.2 | 63.4 |

## Releases

The model is trained in a self-evolving loop, and new versions will follow.

| model | date | DI 0.3 public | notes |
|---|---|---|---|
| KnowLine-35B-A3B-Gen1.5 (this model) | 2026-10-22 | 70.72 | Gen1 weights, read-twice serving |
| [KnowLine-4B-Gen5](https://huggingface.co/PelaAI/KnowLine-4B-Gen5) | 2026-10-21 | 65.67 | 4B, fifth release |
| [KnowLine-26B-A4B-Gen2](https://huggingface.co/PelaAI/KnowLine-26B-A4B-Gen2) | 2026-10-19 | 66.94 | 26B-A4B, second release |
| [KnowLine-35B-A3B-Gen1](https://huggingface.co/PelaAI/KnowLine-35B-A3B-Gen1) | 2026-10-12 | 69.49 | first 35B-A3B release |

## Evaluation

All results are self-run and have not been verified by a third party. Every Gen1.5 number below was measured through the
Gen1.5 front end (read twice on). Unless noted, the Gen1 numbers were measured again on the same files.

### Other public benchmarks

| suite | 35B-A3B-Gen1.5 | 35B-A3B-Gen1 |
|---|---|---|
| C-Eval (4 categories, macro) | 88.1 | 87.7 |
| JevBench / JevBench-hard | 87.0 / 73.0 | 85.7 / 70.3 |

- In a paired test against Gen1, the differences on C-Eval and JevBench have p > 0.05.

### Web operation (Mind2Web official test splits)

| split | Gen1.5 element selection | Gen1.5 operation (balanced) | Gen1 element selection | Gen1 operation (balanced) |
|---|---|---|---|---|
| test_task (Cross-Task) | 95.1 | 97.7 | 94.4 | 97.4 |
| test_website (Cross-Website) | 93.4 | 97.5 | 92.4 | 97.5 |
| test_domain (Cross-Domain) | 92.9 | 98.4 | 93.1 | 98.0 |

- Within noise of Gen1 on all three splits (paired test, p ≥ 0.15); no request failed.

- The task is to pick the target element from a candidate set: the target plus up to 5 other elements sampled from the
  page. This is easier than the original Mind2Web protocol, so do not compare these results with the Mind2Web
  leaderboard.

### Images (zero-shot)

- **Unchanged from Gen1:** read twice applies only to text-only requests. A request with images is sent exactly as in
  Gen1, byte for byte, to the same weights, so Gen1's image results hold for Gen1.5 and we did not rerun them.
- **Gen1's result:** with the official kit, the Decision Index vision public suite gave 73.00 (12,210 rows, complete;
  self-run), published with Gen1's Decision Index results.

### Calibration

Computed on the 0.3 public suite using the method described by the Decision Index board: each field is scored right or
wrong, confidence is the probability of the chosen option, benchmarks are weighted equally, and 10 equal-width bins are
used.

| metric | 35B-A3B-Gen1.5 | 35B-A3B-Gen1 |
|---|---|---|
| ECE (lower is better) | 0.03-0.04 | 0.04-0.05 |
| Brier score (lower is better) | 0.24-0.26 | 0.25-0.27 |
| Confidence ≥95% but wrong | 1.5% | 1.7% |
| Mean confidence / accuracy | 0.85-0.86 / 0.81-0.83 | 0.85-0.86 / 0.80-0.82 |

- **Temperature in the weights:** unchanged from Gen1: the final norm weight includes T = 1.20, fitted on held-out data, so every logit is divided
  by 1.20 before the softmax over the option labels.
- **Why ranges:** the board does not publish which benchmarks it uses for calibration, so we report ranges over the
  plausible benchmark sets, as for Gen1.
- **Recommendation:** if your use depends on probability thresholds, check calibration on your own data.

## Disclosures

- **Training data:** the weights are those of Gen1, so the decontamination statement on the
  [Gen1 card](https://huggingface.co/PelaAI/KnowLine-35B-A3B-Gen1#disclosures) applies unchanged. The training data is
  not included in this repository.
- **Public suite:** the Decision Index public suite is public; the board's private tests may show a lower score.

## Limitations

- **Slower than Gen1:** about 1.56 times the prompt tokens on the Decision Index suite; one request at a time on one H20,
  63.6 ms median and 101 ms mean (Gen1: 51 ms and 73 ms), and 328 ms at the 95th percentile (Gen1: 174 ms). Use
  `--no-read-twice` if latency matters more than these gains.
- **Only short requests change:** text-only requests over 3,000 tokens, and requests with images, are answered exactly as
  by Gen1.
- **Short-text tasks:** read twice does not help everywhere; a few short-text benchmarks are slightly lower than with
  Gen1.
- **Everything else as Gen1:** knowledge-heavy reasoning is still the weakest area (55.0, DI 0.3) and HLE remains at
  chance level; math answers come without step-by-step reasoning; prompt-injection attacks may still work, so keep untrusted
  text clearly delimited and in data fields of the state; the board's private tests may show a lower score than the
  public suite.

## Citation

```bibtex
@misc{knowline35ba3bgen15,
  title  = {KnowLine-35B-A3B-Gen1.5: a 35B-A3B decision model served with read twice},
  author = {PelaAI},
  year   = {2026},
  url    = {https://huggingface.co/PelaAI/KnowLine-35B-A3B-Gen1.5}
}
```

## License

- **Weights:** Apache-2.0, the same as the base model.
- **Code:** `knowline_server.py`, `knowline_layout.py` and `knowline_engine.py` are MIT (see the file headers).