GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
khudgins
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
khudgins
Model Tree
deepreinforce-ai/Ornith-1.0-35B
Input Modalities
Output Modalities
Supported Functionality
This is a LoRA adapter (~130 MB), applied on top of the base MoE deepreinforce-ai/Ornith-1.0-35B
(the base is large — plan for ~70 GB of bf16 weights, or use a quantized base). Want a ready-to-run
quantized build instead? See the GGUF companion repo
khudgins/Ornith-1.0-35B-ThinkingCap-GGUF (Q8_0 recommended).
Quick use — transformers + PEFT:
python
import torchfrom transformers import AutoModelForImageTextToText, AutoTokenizerfrom peft import PeftModelBASE = "deepreinforce-ai/Ornith-1.0-35B"ADAPTER = "khudgins/Ornith-1.0-35B-ThinkingCap" # this repotok = AutoTokenizer.from_pretrained(BASE)model = AutoModelForImageTextToText.from_pretrained(BASE, dtype=torch.bfloat16, device_map="auto")model = PeftModel.from_pretrained(model, ADAPTER).eval()msgs = [{"role": "user", "content": "How many positive integers under 1000 are divisible by neither 5 nor 7?"}]inp = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt",return_dict=True).to(model.device)out = model.generate(**inp, max_new_tokens=1024, do_sample=False)print(tok.decode(out[0][inp["input_ids"].shape[1]:], skip_special_tokens=True))
Hosting — vLLM (LoRA served on the base):
bash
vllm serve deepreinforce-ai/Ornith-1.0-35B \--enable-lora \--lora-modules thinking-cap=khudgins/Ornith-1.0-35B-ThinkingCap \--max-lora-rank 32# then call the OpenAI-compatible API with "model": "thinking-cap"
Greedy decode, in-process (transformers + PEFT), N=50 per benchmark. Each cell is
accuracy @ mean completion tokens.
| Benchmark (N=50) | Base Ornith-35B | Thinking-Cap (ckpt-500) |
|---|---|---|
| GSM8K | 90% @ 720 | 96% @ 546 (−24% tokens, +6 pts) |
| HumanEval | 78% @ 1181 | 78% @ 1125 (held) |
| MBPP | 74% @ 1319 | 76% @ 1284 (+2 pts) |
| average accuracy | 80.7% | 83.3% |
Adaptive concision. The cut scales to how compressible each task is — dramatic on math (reasoning is cheap), near-zero on code (reasoning is load-bearing and the model keeps it). Total token reduction reads modest (~8%) precisely because the code benchmarks dominate the token count and code is deliberately not compressed. Accuracy holds or improves on every benchmark; the checkpoint sweep plateaus and later checkpoints begin to over-compress GSM8K (ckpt-800 drops to 88%), so ckpt-500 — the knee — is the released finalist, not the final step.
Tested on 24 loop-prone prompts (hard reasoning, eval-framing, ambiguous), greedy decode, with a repeated-span detector:
| Model | Loop rate | Runaway completions (hit cap) |
|---|---|---|
| Base Ornith-35B | 0.0% | 4 |
| Thinking-Cap | 0.0% | 1 |
No loops; the tune reduces runaway/truncation on adversarial prompts.
Correctness-gated, length-penalized GRPO (RL with verifiable rewards). Per completion:
markdown
reward = correctness # 1.0 if answer/tests pass, else 0.0− (λ · normalized_length if correct) # length penalty ONLY when correct+ format_bonus
The gate — length penalty applies only to correct answers — makes it impossible to trade accuracy for brevity. GRPO's group-relative advantage means compression happens only on prompts already mastered; a KL leash to the frozen base and a λ ramp keep it stable and self-limiting.
MoE specifics (how this differs from the dense 9B). The base is a 256-expert MoE. LoRA targets
all attention layers (both full-attention and linear-attention/GDN projections, ~32.5M trainable
params); the experts are frozen. This is partly structural: on this architecture the routed
experts are stored as fused parameter tensors (a grouped GEMM), not standard linear layers, so a
stock PEFT LoRA cannot attach to them — it adapts attention and the shared/dense MLP only. Freezing
the experts also preserves the base model's knowledge and improves MoE-RL stability. One consequence:
because "how much to elaborate" is substantially an FFN/expert behavior, steering through attention
alone is a narrower channel than the dense 9B (which did adapt its MLP) — a likely contributor to
the 35B's milder compression. Trained with plain
transformers + PEFT + TRL GRPOTrainer (LoRA r32/α64) on an NVIDIA DGX Spark (GB10). Reward reuses
the same verifiable checkers used to grade the model, so training and eval share one oracle.
MIT. This fine-tune — the LoRA adapter and any weights merged from it — is released under the MIT License. It adds no new base weights; it's a low-rank delta trained on top of the base.
Upstream provenance and terms, stated precisely:
deepreinforce-ai/Ornith-1.0-35B
is tagged MIT on Hugging Face; the inherited base weights are governed by that grant.Qwen3_5MoeForConditionalGeneration).Method: ThinkingCap (correctness-gated, length-penalized GRPO) — an independent reproduction of
the ThinkingCap model series by BottleCap AI
(announcement ·
bottlecapai/ThinkingCap-Qwen3.6-27B),
applied here to a larger Mixture-of-Experts coding model. Not affiliated with or endorsed by
BottleCap AI.
Base: deepreinforce-ai/Ornith-1.0-35B (MIT per its HF tag) → post-trained on Qwen3.5 (Apache 2.0).
Tooling: transformers, PEFT, TRL GRPOTrainer. Full recipe and eval scripts:
github.com/khudgins/ornith-thinking-cap.