--- title: NeoHorse-1-9B canonical_url: "https://www.modelscope.ai/models/TokenRhythm/NeoHorse-1-9B" md_url: "https://www.modelscope.ai/models/TokenRhythm/NeoHorse-1-9B.md" repository: TokenRhythm/NeoHorse-1-9B last_updated: 2026-09-11 license: "Apache License 2.0" pipeline_tag: text-generation tasks: - text-generation model_type: - qwen3_5_text architectures: - Qwen3_5ForCausalLM base_model: - Qwen/Qwen3.5-9B base_model_relation: finetune parameters: 9.0B tensor_type: - BF16 library_name: - safetensors - pytorch frameworks: - Pytorch downloads: 21 stars: 0 tags: - agentic - tool-use - coding - reasoning - instruction-following --- # NeoHorse-1-9B > NeoHorse-1-9B - An open-source model by TokenRhythm on ModelScope. NeoHorse-1-9B Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness. TokenRhythm/NeoHorse-1-9B is a 9.0B-parameter text-generation model on ModelScope. licensed under Apache License 2.0. derived from Qwen/Qwen3.5-9B. - **Repository**: TokenRhythm/NeoHorse-1-9B - **License**: Apache License 2.0 - **Tasks**: text-generation - **Parameters**: 9.0B - **Base model**: Qwen/Qwen3.5-9B - **Tags**: agentic, tool-use, coding, reasoning, instruction-following - **Downloads**: 21 - **Stars**: 0 - **Last updated**: 2026-09-11 Source: https://www.modelscope.ai/models/TokenRhythm/NeoHorse-1-9B ---

NeoHorse-1-9B

Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness.

GitHub ModelScope Hugging Face Company Twitter / X License: Apache-2.0

Technical Report

NeoHorse-1-9B is a 9B causal language model and an initial prototype on the path toward **recursive self-improvement (RSI)**. It is post-trained from Qwen3.5-9B for text-based agent harnesses, tool use, coding, and instruction following. Derived from [Qwen/Qwen3.5-9B](https://huggingface.co/Qwen/Qwen3.5-9B) and fine-tuned by TokenRhythm. This release contains **language-model weights only** and is repackaged for text-only inference. Vision weights are not included. Repackaging changes configuration and tensor key names, without changing the fine-tuned tensor values.

NeoHorse-1-9B evaluation results

## Highlights - **Path toward RSI:** the routing harness assigns tasks to a heterogeneous model pool, records tool interactions and outcomes, estimates capability demand, and uses capability-level feedback to shape the next training mixture. Updated models can return to the harness, closing a prototype evaluation鈥搒election鈥搖pdate loop; extending this loop across successive iterations is the next step toward RSI. - **Agentic post-training framework:** the associated research explores routing-guided curriculum SFT and routing-guided on-policy distillation to turn execution trajectories into training signal while preserving execution and harness context around each response. - **Data quality:** exact and near-duplicate removal, evaluation decontamination, structural validation, six-dimensional semantic evaluation, and subscene-level Scene/Goal/Outcome labeling. - **Broad gains:** 69.04 macro average across ten benchmarks versus 65.60 for Qwen3.5-9B (**+3.44**). ## Model Details
Property Value
Model family NeoHorse Agent-Native Causal Language Model
Parameters Approximately 9B
Base model Qwen3.5-9B
Post-training Routing-guided agentic post-training
Interface Text input and text output
Context length 262,144 natively and extensible up to 1,010,000 tokens.
Weight format / precision Safetensors / BF16
## Evaluation The 9B track compares NeoHorse-1-9B with five representative open-weight baselines: Granite-4.2-8B, Qwen3.5-9B, Ornith-1.5-9B, Gemma-4-12B-it, and Muse-Glimmer-30B. Results cover ten benchmarks and are grouped by capability. Higher is better; `螖` is NeoHorse-1-9B minus Qwen3.5-9B. **Bold** and underline mark the best and second-best results in each benchmark row, respectively; ties share the same formatting.
Benchmark Granite-4.2-8B Qwen3.5-9B Ornith-1.5-9B Gemma-4-12B-it Muse-Glimmer-30B NeoHorse-1-9B 螖 vs Qwen3.5-9B
馃 Agentic
QwenClawBench
37.01
44.04
47.27
43.53
46.11
48.73
+4.69
WorkBuddy Bench
35.07
39.60
29.29
29.65
45.85
40.15
+0.55
PinchBench
56.93
74.55
68.22
58.89
71.35
82.25
+7.70
VitaBench
23.00
31.25
26.75
36.50
48.50
42.25
+11.00
BFCL v4
52.06
64.88
65.03
62.06
53.74
67.43
+2.55
tau2-Bench
62.28
88.04
83.68
59.37
76.64
90.82
+2.78
馃捇 Coding
HumanEval
96.34
92.68
93.90
100.00
98.17
98.17
+5.49
LiveCodeBench v6
72.00
65.14
47.43
73.14
65.71
65.14
+0.00
馃摎 Instruction Following
IFBench
78.00
66.33
40.00
77.67
78.67
66.33
+0.00
IFEval
92.98
89.46
71.35
94.27
93.90
89.09
-0.37
馃搳 Overall
Ten-benchmark average
60.57
65.60
57.29
63.51
67.86
69.04
+3.44
> **Reported protocol:** SGLang v0.5.17 路 `temperature=1.0` 路 `top_p=0.95` 路 `top_k=20` 路 `min_p=0.0` 路 `presence_penalty=1.5` 路 `repetition_penalty=1.0` 路 thinking mode enabled with `enable_thinking=true` and `force_nonempty_content=true`. QwenClawBench, WorkBuddy Bench, and tau2-Bench use three runs; PinchBench and VitaBench use one run; the remaining benchmarks follow their official protocols. VitaBench uses the DeepSeek-V4-Flash simulator and judge. ## Deployment The examples below are for self-hosted deployment from a downloaded local checkpoint. ### Local checkpoint path The examples below assume the checkpoint has already been downloaded to local disk. Set `MODEL_PATH` to the directory containing `config.json`, tokenizer files, and model weights. ```bash MODEL_PATH="/path/to/NeoHorse-1-9B" ``` The OpenAI-compatible requests below use the server's `--served-model-name` (for example, `neohorse-1-9b`), not the filesystem path. ### SGLang The technical report uses SGLang v0.5.17. ```bash pip install "sglang==0.5.17" MODEL_PATH="/path/to/NeoHorse-1-9B" python3 -m sglang.launch_server \ --model-path "$MODEL_PATH" \ --served-model-name neohorse-1-9b \ --host 0.0.0.0 \ --port 30000 \ --context-length 262144 \ --reasoning-parser qwen3 \ --tool-call-parser qwen3_coder ``` Send an OpenAI-compatible request after the server starts: ```bash curl http://localhost:30000/v1/chat/completions \ -H 'Content-Type: application/json' \ -d '{"model":"neohorse-1-9b","messages":[{"role":"user","content":"Write a Python function that returns the first n Fibonacci numbers."}],"max_tokens":512}' ``` ### vLLM ```bash pip install -U vllm MODEL_PATH="/path/to/NeoHorse-1-9B" vllm serve "$MODEL_PATH" \ --served-model-name neohorse-1-9b \ --host 0.0.0.0 \ --port 8000 \ --max-model-len 262144 \ --reasoning-parser qwen3 \ --enable-auto-tool-choice \ --tool-call-parser qwen3_coder ``` The server exposes an OpenAI-compatible `/v1/chat/completions` endpoint. Send a request after the server starts: ```bash curl http://localhost:8000/v1/chat/completions \ -H 'Content-Type: application/json' \ -d '{"model":"neohorse-1-9b","messages":[{"role":"user","content":"Write a Python function that returns the first n Fibonacci numbers."}],"max_tokens":512}' ``` The example uses the configured 262,144-token context limit. Actual capacity depends on GPU memory and serving settings; reduce the context limit if needed. These launch examples have not yet been validated on GPU for this repackaged release. ## License NeoHorse-1-9B is released under the **Apache License 2.0**. The upstream model is [Qwen/Qwen3.5-9B](https://huggingface.co/Qwen/Qwen3.5-9B). Its original copyright notice, Copyright 2026 Alibaba Cloud, is retained in the license file. TokenRhythm has modified the model through fine-tuning and repackaging for text-only inference. Modification notices are included in this model card and the released configuration, weight index, and Safetensors metadata. ## Citation ``` @misc{neohorse2026, title = {NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness}, author = {NeoHorse Team}, year = {2026}, howpublished = {arXiv preprint}, eprint = {2609.08183}, archivePrefix = {arXiv}, primaryClass = {cs.CL}, url = {https://arxiv.org/abs/2609.08183} } ``` For questions or issue reports, use the [NeoHorse project repository](https://github.com/TokenRhythm/NeoHorse).