GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Arun63
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Arun63
Model Tree
Qwen/Qwen3.6-35B-A3B
Input Modalities
Output Modalities
Supported Functionality
The MMLU-Pro pass used 70 total questions per model: --limit 5 across 14 MMLU-Pro subjects. Treat this as a smoke/comparative check, not a release-quality full benchmark.
| Benchmark | Harness | Samples per model | Setting | Metric | Base model | Fine-tuned merged model | Delta |
|---|---|---|---|---|---|---|---|
| MMLU-Pro overall | lm-evaluation-harness | 70 | --limit 5 across 14 subjects | exact_match, custom-extract | 42.86% |
| 75.71% |
| +32.85 pp |
Base model: Qwen/Qwen3.6-35B-A3B. Fine-tuned model: hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled.
[!WARNING] Community benchmarks welcome
To better understand this fine-tuned model's capabilities, I welcome independent benchmark results. If you run evaluations, please include the benchmark name, harness/script, sample count, decoding settings, and raw logs or result files when possible.
Share results by opening a PR/discussion or DMing @hesamation on X.
This release delivers substantial upgrades, particularly in:
For more details, please refer to the Qwen blog post Qwen3.6-35B-A3B.
The following table is from the upstream Qwen3.6-35B-A3B release context and is included for base-model reference. It is not a benchmark of this fine-tuned checkpoint unless explicitly stated in the fine-tune benchmark table above.
| Category | Benchmark | Qwen3.5-27B | Gemma4-31B | Qwen3.5-35BA3B | Gemma4-26BA4B | Qwen3.6-35BA3B |
|---|---|---|---|---|---|---|
| Coding Agent | SWE-bench Verified | 75.0 | 52.0 | 70.0 | 17.4 | 73.4 |
| Coding Agent | SWE-bench Multilingual | 69.3 | 51.7 | 60.3 | 17.3 | 67.2 |
| Coding Agent | SWE-bench Pro | 51.2 | 35.7 | 44.6 | 13.8 | 49.5 |
| Coding Agent | Terminal-Bench 2.0 | 41.6 | 42.9 | 40.5 | 34.2 | 51.5 |
| Coding Agent | Claw-Eval Avg | 64.3 | 48.5 | 65.4 | 58.8 | 68.7 |
| Coding Agent | Claw-Eval Pass^3 | 46.2 | 25.0 | 51.0 | 28.0 | 50.0 |
| Coding Agent | SkillsBench Avg5 | 27.2 | 23.6 | 4.4 | 12.3 | 28.7 |
| Coding Agent | QwenClawBench | 52.2 | 41.7 | 47.7 | 38.7 | 52.6 |
| Coding Agent | NL2Repo | 27.3 | 15.5 | 20.5 | 11.6 | 29.4 |
| Coding Agent | QwenWebBench | 1068 | 1197 | 978 | 1178 | 1397 |
| General Agent | TAU3-Bench | 68.4 | 67.5 | 68.9 | 59.0 | 67.2 |
| General Agent | VITA-Bench | 41.8 | 43.0 | 29.1 | 36.9 | 35.6 |
| General Agent | DeepPlanning | 22.6 | 24.0 | 22.8 | 16.2 | 25.9 |
| General Agent | Tool Decathlon | 31.5 | 21.2 | 28.7 | 12.0 | 26.9 |
| General Agent | MCPMark | 36.3 | 18.1 | 27.0 | 14.2 | 37.0 |
| General Agent | MCP-Atlas | 68.4 | 57.2 | 62.4 | 50.0 | 62.8 |
| General Agent | WideSearch | 66.4 | 35.2 | 59.1 | 38.3 | 60.1 |
| Knowledge | MMLU-Pro | 86.1 | 85.2 | 85.3 | 82.6 | 85.2 |
| Knowledge | MMLU-Redux | 93.2 | 93.7 | 93.3 | 92.7 | 93.3 |
| Knowledge | SuperGPQA | 65.6 | 65.7 | 63.4 | 61.4 | 64.7 |
| Knowledge | C-Eval | 90.5 | 82.6 | 90.2 | 82.5 | 90.0 |
| STEM & Reasoning | GPQA | 85.5 | 84.3 | 84.2 | 82.3 | 86.0 |
| STEM & Reasoning | HLE | 24.3 | 19.5 | 22.4 | 8.7 | 21.4 |
| STEM & Reasoning | LiveCodeBench v6 | 80.7 | 80.0 | 74.6 | 77.1 | 80.4 |
| STEM & Reasoning | HMMT Feb 25 | 92.0 | 88.7 | 89.0 | 91.7 | 90.7 |
| STEM & Reasoning | HMMT Nov 25 | 89.8 | 87.5 | 89.2 | 87.5 | 89.1 |
| STEM & Reasoning | HMMT Feb 26 | 84.3 | 77.2 | 78.7 | 79.0 | 83.6 |
| STEM & Reasoning | IMOAnswerBench | 79.9 | 74.5 | 76.8 | 74.3 | 78.9 |
| STEM & Reasoning | AIME26 | 92.6 | 89.2 | 91.0 | 88.3 | 92.7 |
Notes from the upstream Qwen3.6 release:
text
Qwen/Qwen3.6-35B-A3B-> supervised fine-tuning with LoRA-> merged full model-> Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled
Training configuration:
| Setting | Value |
|---|---|
| Fine-tuning method | Supervised fine-tuning with LoRA |
| LoRA target | Attention-only modules |
| LoRA rank / alpha | 32 / 32 |
| Micro-batch size | 1 |
| Gradient accumulation | 32 |
| Epochs | 2 |
| Completed steps | 762 / 762 |
| Final reported training loss | 0.3362497625740494 |
| Dataset max tokens | 8192 |
| Max sequence length | 32768 |
The recipe samples and normalizes reasoning conversations from three datasets, then renders them with the qwen3-thinking chat template and response-only SFT masking.
| Dataset | Requested sample count | Role |
|---|---|---|
nohurry/Opus-4.6-Reasoning-3000x-filtered | 3,900 | Claude Opus reasoning trajectories |
Jackrong/Qwen3.5-reasoning-700x | 700 | Curated Qwen reasoning samples |
Roman1111111/claude-opus-4.6-10000x | 9,633 | Additional Claude Opus reasoning examples |
This model is intended for reasoning-heavy text workflows such as coding assistance, planning, math-style reasoning, and structured analytical responses. Because the fine-tune is text-only, image/video behavior should be treated as inherited from the base model rather than improved by this training run.