GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
jihwan1205
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
jihwan1205
Model Tree
ModalityDance/latent-tts-coconut
Input Modalities
Output Modalities
Supported Functionality
Clean greedy decoding, max_new_tokens=64, exact-match on the final answer.
| Model (GPT-2 124M) | GSM8K | GSM-Hard | MultiArith | SVAMP | ASDiv-A | GSM-Plus |
|---|---|---|---|---|---|---|
| COCONUT (base) | 34.1 | 7.7 | 80.9 | 35.6 | 60.2 | 17.4 |
| SLPO | 34.9 | 7.6 | 82.8 | 34.3 | 58.7 | 18.2 |
| SIM-CoT | 44.7 | 9.3 | 90.5 | 40.6 | 67.2 | 21.5 |
| CoDi | 42.5 | 9.3 | 91.9 | 40.0 | 65.4 | 23.1 |
This model (SVP on W_V, α=0.6) | 50.3 | 11.3 | 93.4 | 43.6 | 71.2 | 27.6 |
This is a fixed-length continuous-thought model: the prompt ends with
<|start-latent|>, six latent steps feed each step's last hidden state back as the next
input embedding, <|end-latent|> closes the latent phase, and the answer is then decoded
as ordinary tokens.
python
from transformers import AutoTokenizer, GPT2LMHeadModel REPO = "." # path to this repositorytok = AutoTokenizer.from_pretrained(REPO)model = GPT2LMHeadModel.from_pretrained(REPO).eval().cuda()# then run the two-phase latent loop: 6 latent steps, then decode the answerA reference implementation of the loop ships with the SVP code release; the original COCONUT inference code works as well.
| Base | ModalityDance/latent-tts-coconut (COCONUT GPT-2, 124M) |
| Data | 383,891 eligible prompts from GSM8K-Aug |
| Algorithm | GRPO: G=32 rollouts per prompt; 2B=64 candidate prompts per iteration with up to B=32 mixed-outcome groups retained; μ=2 optimization passes; centered advantages; symmetric clipping ε=0.2; k3 KL (β=0.02) to the frozen base |
| Exploration |
MIT, inherited from the base checkpoint (which derives from openai-community/gpt2).
SVP with α=0.6: independent Gaussian coefficient perturbations for each Value-projection layer and rollout, held fixed through prefill and all six latent steps; clean answer decoding; SVD factors refreshed every 50 iterations |
| Trained parameters | V columns of every c_attn (≈ 7M of 124M) |
| Optimizer | AdamW, constant lr=6e-5, seed 0, no warmup or weight decay, gradient clipping at 1.0; epoch-9 checkpoint from 15 epochs |
| Reward | Final-answer correctness only |