Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more
Container

Run this model inference with full control and performance in your environment.

Learn more

Get help setting up a custom Dedicated Endpoints.

Talk with our engineer to get a quote for reserved GPU instances with discounts.

README

License: apache-2.0

Model details

ArchitectureLlama-style decoder (RoPE, RMSNorm, SwiGLU, GQA)
Parameters1.246 B (tied input/output embeddings)
Hidden / layers2048 / 16
Attention heads32 query / 8 KV (GQA)
Intermediate5632
Context length2048
Vocab256 000 (morpheme-BPE)
Precisionbf16

Training

Tokens6.26 B (1 epoch)
Train blocks3 057 865 × 2048
Corpuscleaned → MinHash-deduped (11.29 M / 13.19 M docs kept, 85.6 %)
Hardware8 × NVIDIA H200, FSDP full-shard, bf16
OptimizerAdamW (β 0.9/0.95, wd 0.1), cosine LR 3e-4, warmup 200
Effective batch512 blocks (8 × 16 × grad-accum 4) ≈ 1.05 M tok/step
Throughput~313 K tok/s
Wall-clock~5 h 40 m
Final loss~2.90 (train)

Usage

python

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
name = "stukenov/Til-Core-1B"
tok = AutoTokenizer.from_pretrained(name)
m = AutoModelForCausalLM.from_pretrained(name, dtype=torch.bfloat16).cuda().eval()
ids = tok("Қазақстан Республикасы — ", return_tensors="pt").input_ids.cuda()
out = m.generate(ids, max_new_tokens=50, do_sample=True,
temperature=0.8, top_p=0.95, repetition_penalty=1.2)
print(tok.decode(out[0], skip_special_tokens=True))

Sample generations

Қазақстан Республикасы — мемлекеттік рәміздері. Жалпы білім беретін мектептің 6-сыныбына арналған оқулық…

Жасанды интеллект дегеніміз бұл адам миының эволюциясы, ойлау жүйесі мен мінез-құлқының ерекшеліктерін…

Менің Отаным — «Отан» туралы өлеңді мәнерлеп оқу… Біздің Отанымыз қалай аталады?…

Limitations

  • Base model — no instruction following, no safety alignment.
  • Single epoch on a 6.26 B-token corpus; factual reliability is limited.
  • Corpus skews toward educational / encyclopedic Kazakh text; occasional rare-token artifacts in generation.
  • Kazakh-centric; not optimized for other languages.

Roadmap

  • Til Core 1B Instruct — SFT on Kazakh instruction data (see plan in repo).
  • A smaller instruct sibling for on-device use.

Citation

markdown

@misc{tilcore1b2026,
title = {Til Core 1B: a Kazakh base language model with a morpheme-BPE tokenizer},
author = {Tukenov, Saken},
year = {2026}
}

Model provider

TilQazyna

TilQazyna

Model tree

Base

this model

Modalities

Input

Text

Output

Text

Pricing

Dedicated Endpoints

View details

Supported Functionality

Model APIs

Dedicated Endpoints

Container

More information

Explore FriendliAI today