Papers › Chronicals: A High-Performance Framework for LLM Fine-Tuning with 3.51x Speedup over Unsloth

Chronicals: A High-Performance Framework for LLM Fine-Tuning with 3.51x Speedup over Unsloth

6 Jan 2026arXiv:2601.02609added by Syntology

Arjun S. Nair

Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).

Large language model fine-tuning is bottlenecked by memory: a 7B parameter model requires 84GB--14GB for weights, 14GB for gradients, and 56GB for FP32 optimizer states--exceeding even A100-40GB capacity. We present Chronicals, an open-source training framework achieving 3.51x speedup over Unsloth through four synergistic optimizations: (1) fused Triton kernels eliminating 75% of memory traffic via RMSNorm (7x), SwiGLU (5x), and QK-RoPE (2.3x) fusion; (2) Cut Cross-Entropy reducing logit memory from 5GB to 135MB through online softmax computation; (3) LoRA+ with theoretically-derived 16x differential learning rates between adapter matrices; and (4) Best-Fit Decreasing sequence packing recovering 60-75% of compute wasted on padding. On Qwen2.5-0.5B with A100-40GB, Chronicals achieves 41,184 tokens/second for full fine-tuning versus Unsloth's 11,736 tokens/second (3.51x). For LoRA at rank 32, we reach 11,699 tokens/second versus Unsloth MAX's 2,857 tokens/second (4.10x). Critically, we discovered that Unsloth's reported 46,000 tokens/second benchmark exhibited zero gradient norms--the model was not training. We provide complete mathematical foundations: online softmax correctness proofs, FlashAttention IO complexity bounds O(N^2 d^2 M^{-1}), LoRA+ learning rate derivations from gradient magnitude analysis, and bin-packing approximation guarantees. All implementations, benchmarks, and proofs are available at https://github.com/Ajwebdevs/Chronicals with pip installation via https://pypi.org/project/chronicals/.

PaperPDFCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 33 code samples harvested from 1 repository linked to this paper: 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 26 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: found in paper text by Syntology: 33 samples from 1 repository, 7 ran (5 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

Ajwebdevs/Chronicals found in paper text by SyntologyNOASSERTIONSyntology: 33 samples, 7 ran (2 instrument-failed) · 26 unverified report

Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

33 samples harvested: 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 26 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · our draft was wrong
5ran
26unverified

Licence: 33 of the 33 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from Ajwebdevs/Chronicals. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

Ajwebdevs/Chronicals found in paper text by Syntology 33 samples, 7 ran (2 instrument-failed) · 26 unverified
chunked_cross_entropy_loss chronicals/kernels/cut_cross_entropy.py ran licence not identified · pointer only · 549a700a723e8ef2 · report fixture: signature logits:Tensor/3/float, labels:Tensor/2/int, chunk_size:int/0/int, ignore_index:int/0/int, label_smoothing:float/0/float, z_loss_weight:float/0/float · returned Tensor · level recorded 2026-09-26
compute_arithmetic_intensity chronicals/kernels/flash_attention_optimizer.py ran licence not identified · pointer only · abab4a1b18037b93 · report fixture: signature seq_len:int/0/int, head_dim:int/0/int, num_heads:int/0/int, batch_size:int/0/int · returned dict · level recorded 2026-09-26
estimate_checkpointing_savings chronicals/lora/gradient_checkpointing_lora.py ran licence not identified · pointer only · 39bad39ba513a138 · report fixture: signature num_layers:int/0/int, hidden_size:int/0/int, seq_length:int/0/int, batch_size:int/0/int, mode:str/0/str · returned dict · level recorded 2026-09-26
estimate_tokens_per_second chronicals/kernels/flash_attention_optimizer.py ran licence not identified · pointer only · 7b8b337acd50fdf4 · report fixture: signature num_params:int/0/int, batch_size:int/0/int, seq_len:int/0/int, gpu_tflops:float/0/float, mfu:float/0/float · returned float · level recorded 2026-09-26
get_gpu_specs chronicals/kernels/flash_attention_optimizer.py ran · our draft was wrong fingerprinted licence not identified · pointer only · e0fee5939a703ab5 · report fixture: signature device_name:str/0/str · returned dict · level recorded 2026-09-26
get_model_size in 21 papers chronicals/lora/lora_presets.py ran · our draft was wrong fingerprinted licence not identified · pointer only · c8e820f2fc086863 · report fixture: signature model_name:str/0/str · returned str · level recorded 2026-09-26
get_optimal_chunk_size chronicals/kernels/cut_cross_entropy.py ran licence not identified · pointer only · c91de53c313dfbb4 · report fixture: signature vocab_size:int/0/int, hidden_dim:int/0/int, available_vram_gb:float/0/float · returned int · level recorded 2026-09-26
apply_dora_to_model chronicals/lora/dora_adapter.py unverified licence not identified · pointer only · 3c9010c06cd4eafc · report
apply_lora_checkpointing chronicals/lora/gradient_checkpointing_lora.py unverified licence not identified · pointer only · 37f4da134de23ea0 · report
apply_lora_to_model chronicals/lora/lora_optimized.py unverified licence not identified · pointer only · def82cf7461704a9 · report
convert_lora_to_dora chronicals/lora/dora_adapter.py unverified licence not identified · pointer only · 125e8438c36c3122 · report
cut_cross_entropy chronicals/kernels/cut_cross_entropy.py unverified licence not identified · pointer only · 022a2eebc3a6077f · report
fused_cross_entropy_triton chronicals/kernels/triton_kernels.py unverified licence not identified · pointer only · 754961defcfd7acd · report
fused_lora_down_dropout chronicals/kernels/fused_lora_kernels.py unverified licence not identified · pointer only · 8e03397a90e03548 · report
fused_lora_linear chronicals/kernels/fused_lora_kernels.py unverified licence not identified · pointer only · c90946ad3d56b68b · report
fused_lora_linear_with_grad chronicals/kernels/fused_lora_kernels.py unverified licence not identified · pointer only · 86fb35acea6bf0e1 · report
fused_q_rope_inplace chronicals/kernels/fused_rope.py unverified licence not identified · pointer only · 0776674b13b98da3 · report
fused_qk_rope chronicals/kernels/fused_qk_rope.py unverified licence not identified · pointer only · a4924b02da2cb36b · report
fused_qk_rope_backward chronicals/kernels/fused_qk_rope.py unverified licence not identified · pointer only · 38fc5ac3f8050ea8 · report
fused_qk_rope_forward chronicals/kernels/fused_qk_rope.py unverified licence not identified · pointer only · 7fa6f1e0f24b4896 · report
fused_qk_rope_inplace chronicals/kernels/fused_rope.py unverified licence not identified · pointer only · b6d424583f8275e7 · report
fused_qk_rope_inplace_backward chronicals/kernels/fused_rope.py unverified licence not identified · pointer only · 9e1e207e9d289234 · report
get_dora_plus_param_groups chronicals/lora/dora_adapter.py unverified licence not identified · pointer only · 55495c41526b01fb · report
get_lora_optimizer_groups chronicals/lora/lora_optimized.py unverified licence not identified · pointer only · 3b1cee1666f8fdb7 · report
get_model_config in 45 papers chronicals/config/config.py unverified licence not identified · pointer only · b2d1236317b593ab · report
get_optimal_config_for_gpu chronicals/config/config.py unverified licence not identified · pointer only · 18e9aa9520439eb6 · report
get_optimal_preset chronicals/lora/lora_presets.py unverified licence not identified · pointer only · 2c25455de3cd069c · report
get_preset_for_hardware chronicals/lora/lora_presets.py unverified licence not identified · pointer only · ff5237ff47007036 · report
get_sota_config_for_gpu chronicals/config/sota_config.py unverified licence not identified · pointer only · 147e044504ca0fd5 · report
print_comprehensive_diagnostic chronicals/config/sota_config.py unverified licence not identified · pointer only · b771b43e585c9731 · report
print_diagnostic_report chronicals/config/sota_config.py unverified licence not identified · pointer only · 669f295dc42160cf · report
rmsnorm_backward chronicals/kernels/triton_kernels.py unverified licence not identified · pointer only · 1be8da1291321955 · report
rmsnorm_forward chronicals/kernels/triton_kernels.py unverified licence not identified · pointer only · f985053a65ebe136 · report

Results from the paper

The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2601.02609, was issued in January 2026, after that date, so the archive has no leaderboard rows for it.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections