GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
tbilisi-ai-lab
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
tbilisi-ai-lab
Model Tree
tbilisi-ai-lab/kona2-12B-Instruct
Input Modalities
Output Modalities
Supported Functionality
| Property | Value |
|---|---|
| Parameters | 12B |
| Architecture | Mistral (Transformer) |
| Context Length | 32K tokens |
| Languages | Georgian (ka), English (en), other (limited) |
| Training | Direct Preference Optimization (DPO) |
| Preference Pairs | 387,319 |
| Function Calling | Yes (Hermes format) |
| Base Model | kona2-12B-Instruct |
text
mistralai/Mistral-Nemo-Base-2407 │ ├── Expand Vocabulary (+20K Georgian tokens) │ └── kona2-12B-Base (continue pre-training, ~30B tokens) │ └── kona2-12B-Instruct (SFT on ~2.8M instructions) │ └── kona2-12B (DPO on 387K preference pairs) ← YOU ARE HERE| Dataset | Pairs | Description |
|---|---|---|
| kona-dpo-mix-387k | 387,319 | Mixed preference pairs |
DPO Pair Sources:
| Source | Pairs | Strategy |
|---|---|---|
| Instruction Augmentation (Sonnet 3.5) | ~29K | Claude 3.5 Sonnet as "chosen", base model as "rejected" |
| Function Calling (tools+call) | ~5K | Correct tool calls vs text responses |
| Function Calling (tools, no call) | ~5K | Correct text vs incorrect tool calls |
| Function Calling (no tools) | ~5K | Appropriate responses without tools |
| Hermes Chats | ~17K | High-quality conversation pairs |
| General Instructions | ~320K | Mixed instruction following |
The model was trained on 4 distinct function-calling scenarios:
| Scenario | Tools Available | Should Call | Chosen | Rejected |
|---|---|---|---|---|
system_tools_yes_call_yes | Yes | Yes | Tool call | Text response |
system_tools_yes_call_no | Yes | No | Text response | Tool call |
tools_yes_call_yes | Yes | Yes | Tool call | Text response |
tools_yes_call_no | Yes | No | Text response | Tool call |
bash
pip install transformers torch acceleratepython
from transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained( "tbilisi-ai-lab/kona2-12B", torch_dtype="auto", device_map="auto")tokenizer = AutoTokenizer.from_pretrained("tbilisi-ai-lab/kona2-12B") messages = [ {"role": "system", "content": "You are a helpful, harmless, and honest assistant."}, {"role": "user", "content": "დამეხმარე პითონზე ფუნქციის დაწერაში, რომელიც ითვლის ფაქტორიალს."}] inputs = tokenizer.apply_chat_template( messages, return_tensors="pt", add_generation_prompt=True).to(model.device) outputs = model.generate( inputs, max_new_tokens=512, temperature=0.7, do_sample=True)print(tokenizer.decode(outputs[0], skip_special_tokens=True))| Model | Stage | Use Case |
|---|---|---|
| kona2-12B-Base | Pre-training | Further fine-tuning |
| kona2-12B-Instruct | SFT | If DPO behavior is not desired |
| kona2-small-3.8B | All | Resource-constrained environments |
bibtex
@misc{tbilisi2025kona2, title = {Kona2-12B: A DPO-Aligned Georgian Language Model}, author = {Tbilisi AI Lab Team}, year = {2025}, publisher = {Hugging Face}, howpublished = {\url{https://huggingface.co/tbilisi-ai-lab/kona2-12B}}}This model is released under the Apache 2.0 License.
mistralai/Mistral-Nemo-Base-2407 │ ├── Expand Vocabulary (+20K Georgian tokens) │ └── kona2-12B-Base (continue pre-training, ~30B tokens) │ └── kona2-12B-Instruct (SFT on ~2.8M instructions) │ └── kona2-12B (DPO on 387K preference pairs) ← YOU ARE HEREfrom transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained( "tbilisi-ai-lab/kona2-12B", torch_dtype="auto", device_map="auto")tokenizer = AutoTokenizer.from_pretrained("tbilisi-ai-lab/kona2-12B") messages = [ {"role": "system", "content": "You are a helpful, harmless, and honest assistant."}, {"role": "user", "content": "დამეხმარე პითონზე ფუნქციის დაწერაში, რომელიც ითვლის ფაქტორიალს."}] inputs = tokenizer.apply_chat_template( messages, return_tensors="pt", add_generation_prompt=True).to(model.device) outputs = model.generate( inputs, max_new_tokens=512, temperature=0.7, do_sample=True)print(tokenizer.decode(outputs[0], skip_special_tokens=True))@misc{tbilisi2025kona2, title = {Kona2-12B: A DPO-Aligned Georgian Language Model}, author = {Tbilisi AI Lab Team}, year = {2025}, publisher = {Hugging Face}, howpublished = {\url{https://huggingface.co/tbilisi-ai-lab/kona2-12B}}}