SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.00835 · ICML · 2024

CLLMs: Consistency Large Language Models

Zhijie Deng, Hao Zhang, Zhezhi He, Siqi Kou, Lanxiang Hu

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2403.00835.json · badge: /api/ran/2403.00835.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
hao-ai-lab/consistency_llm canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
jacobi_generate Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hao-ai-lab/consistency_llm/applications/chat_cli_cllm.py
code served (permissive licence) · get_code("db6e98f9aedf9a96")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned tuple. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import os

torch.manual_seed(0)

# Create a minimal tokenizer stand-in
class MinimalTokenizer:
    def __init__(self):
        self.eos_token_id = 2
        self.pad_token_id = 0

# Create a minimal model with jacobi_forward method
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.embed = nn.Embedding(10, 4)
        self.lm_head = nn.Linear(4, 10)
    
    def jacobi_forward(self, input_ids, tokenizer, max_new_tokens, past_key_values, use_cache, prefill_phase, chat):
        # Returns: (n_gram_generation, first_correct_token, iter_steps, accurate_length)
        bsz = input_ids.shape[0]
        seq_len = input_ids.shape[1]
        
        if prefill_phase:
            # First call: return past_key_values and first_correct_token
            first_correct_token = torch.randint(0, 10, (bsz, 1))
            past_key_values = {"cache": torch.randn(bsz, 4, 4)}
            return past_key_values, first_correct_token
        else:
            # Subsequent calls: return generation, first_correct_token, iter_steps, accurate_length
            n_gram_generation = torch.randint(0, 10, (bsz, max_new_tokens))
            first_correct_token = torch.randint(0, 10, (bsz, 1))
            iter_steps = 2
            accurate_length = 3
            return n_gram_generation, first_correct_token, iter_steps, accurate_length

# Set up environment
os.environ['CHAT'] = '0'

# Create instances
tokenizer = MinimalTokenizer()
model = MinimalModel()

# Create inputs
inputs = {
    'input_ids': torch.randint(0, 10, (1, 5)),
    'attention_mask': torch.ones(1, 5)
}

max_new_tokens = 3
max_new_seq_len = 20

ARGS = {
    'inputs': inputs,
    'model': model,
    'tokenizer': tokenizer,
    'max_new_tokens': max_new_tokens,
    'max_new_seq_len': max_new_seq_len
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Parallel decoding methods such as Jacobi decoding show promise for more efficient LLM inference as it breaks the sequential nature of the LLM decoding process and transforms it into parallelizable computation. However, in practice, it achieves little speedup compared to traditional autoregressive (AR) decoding, primarily because Jacobi decoding seldom accurately predicts more than one token in a single fixed-point iteration step. To address this, we develop a new approach aimed at realizing fast convergence from any state to the fixed point on a Jacobi trajectory. This is accomplished by refining the target LLM to consistently predict the fixed point given any state as input. Extensive experiments demonstrate the effectiveness of our method, showing 2.4× to 3.4× improvements in generation speed while preserving generation quality across both domain-specific and open-domain benchmarks. Our code is available at https://github.com/hao-ailab/Consistency LLM.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2403.00835")
get_code_for_paper("2403.00835")
have("2403.00835")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2403.00835.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2403.00835.svg)](https://syntology.ai/paper/2403.00835)

Connect an agent — have() is free.