GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
wenzani
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
wenzani
Model Tree
Qwen/Qwen3.5-2B
Input Modalities
Output Modalities
Supported Functionality
Deploy with a recent SGLang or vLLM build supporting Qwen3.5. Install the engine for your GPU platform using its linked installation guide, then run one of the commands below from the downloaded model directory. Replace . with wenzani/Wendi-2B to load from Hugging Face instead.
bash
python -m sglang.launch_server \ --model-path . \ --served-model-name wendi-2b \ --dtype bfloat16 \ --tp-size 1 \ --host 127.0.0.1 \ --port 8000Installation guide · Qwen3.5 serving guide
bash
vllm serve . \ --served-model-name wendi-2b \ --dtype bfloat16 \ --tensor-parallel-size 1 \ --language-model-only \ --host 127.0.0.1 \ --port 8000Both expose an OpenAI-compatible API at http://127.0.0.1:8000/v1, with model name wendi-2b. For decision requests, set chat_template_kwargs to {"enable_thinking": false}, use uppercase option letters, and score the next token over the allowed alternatives. The configured context is 262,144 tokens; adjust SGLang's --context-length or vLLM's --max-model-len to fit your available memory.
Internal held-out evaluations of Wendi 2B in BF16, using Transformers and restricted next-token option scoring on an AMD Radeon Pro W7900. All values below are accuracy; counts show correct answers out of evaluated examples. These measurements use the evaluation harness, not the serving engines above.
| Category | Correct / examples | Accuracy |
|---|---|---|
| Banking intent | 131 / 160 | 81.88% |
| Contract evidence | 101 / 160 | 63.12% |
| Emotion classification | 122 / 160 | 76.25% |
| Multilingual intent | 123 / 180 | 68.33% |
| Numeric reasoning | 48 / 100 | 48.00% |
| Qualitative reasoning | 94 / 100 |
| Category | Correct / examples | Accuracy |
|---|---|---|
| Event order | 57 / 60 | 95.00% |
| Latency checks | 57 / 60 | 95.00% |
| Token-count metadata | 53 / 60 | 88.33% |
| Tool-call detection | 79 / 102 | 77.45% |
| Tool-call counting | 74 / 99 | 74.75% |
| Exact user quotations | 73 / 99 |
| Category | Correct / examples | Accuracy |
|---|---|---|
| Natural-email phishing | 135 / 160 | 84.38% |
| Synthetic phishing contrasts | 115 / 120 | 95.83% |
| Spam classification | 76 / 80 | 95.00% |
Two internal holdouts: the established suite has 120 examples per category; the additional scenario suite has 40 per category.
| Category | Established suite | Additional scenarios |
|---|---|---|
| Constraint selection | 66.67% | 77.50% |
| Evidence scope | 93.33% | 92.50% |
| Exception policies | 86.67% | 90.00% |
| Numeric policies | 62.50% | 57.50% |
| Rule precedence | 74.17% | 87.50% |
| Untrusted-content handling | 99.17% |
Wendi was trained with supervised fine-tuning followed by hybrid GRPO, combining decision rewards, supervised loss and KL regularization. Language-only rank-16 LoRA adapters were merged into the BF16 checkpoint.
The final training stage used 4,000 examples over 1,000 optimizer updates:
| Component | Examples |
|---|---|
| General-decision replay | 1,600 |
| Behavior analysis | 800 |
| Email classification | 800 |
| Instruction following | 800 |
Data combines published datasets, synthetic tool-use conversations and original procedural tasks. Sources include:
Apache-2.0. Wendi is a modified derivative of Qwen3.5-2B, incorporating supervised and hybrid-RL language-model updates. The upstream license is included in LICENSE. Training sources retain their original licenses and attribution.
| 94.00% |
| State tracking | 93 / 100 | 93.00% |
| Temporal and numeric reasoning | 33 / 60 | 55.00% |
| Workflow rules | 70 / 100 | 70.00% |
| Overall | 815 / 1,120 | 72.77% |
| 73.74% |
| Overall | 393 / 480 | 81.88% |
| 100.00% |
| Overall | 80.42% (579 / 720) | 84.17% (202 / 240) |