Instructions to use AIIT-Threshold/Tessera-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AIIT-Threshold/Tessera-1B with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="AIIT-Threshold/Tessera-1B",
	filename="gguf/tessera-1b-Q6_K.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use AIIT-Threshold/Tessera-1B with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf AIIT-Threshold/Tessera-1B:Q6_K
# Run inference directly in the terminal:
llama cli -hf AIIT-Threshold/Tessera-1B:Q6_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf AIIT-Threshold/Tessera-1B:Q6_K
# Run inference directly in the terminal:
llama cli -hf AIIT-Threshold/Tessera-1B:Q6_K

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf AIIT-Threshold/Tessera-1B:Q6_K
# Run inference directly in the terminal:
./llama-cli -hf AIIT-Threshold/Tessera-1B:Q6_K

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf AIIT-Threshold/Tessera-1B:Q6_K
# Run inference directly in the terminal:
./build/bin/llama-cli -hf AIIT-Threshold/Tessera-1B:Q6_K

Use Docker

docker model run hf.co/AIIT-Threshold/Tessera-1B:Q6_K

LM Studio
Jan
Ollama
How to use AIIT-Threshold/Tessera-1B with Ollama:
```
ollama run hf.co/AIIT-Threshold/Tessera-1B:Q6_K
```

Unsloth Studio

How to use AIIT-Threshold/Tessera-1B with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for AIIT-Threshold/Tessera-1B to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for AIIT-Threshold/Tessera-1B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for AIIT-Threshold/Tessera-1B to start chatting

Atomic Chat new
Docker Model Runner
How to use AIIT-Threshold/Tessera-1B with Docker Model Runner:
```
docker model run hf.co/AIIT-Threshold/Tessera-1B:Q6_K
```

Lemonade

How to use AIIT-Threshold/Tessera-1B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull AIIT-Threshold/Tessera-1B:Q6_K

Run and chat with the model

lemonade run user.Tessera-1B-Q6_K

List all available models

lemonade list

Tessera-1B / config.json

AIIT-Threshold

Tessera 1B: from-scratch 1.01B base + v12i/v7 SFT adapters, tokenizer, loader, USAGE (Apache-2.0)

abfb518 verified 18 days ago

Raw

History Blame Contribute Delete

1.18 kB

	{
	"model_family": "tessera",
	"architecture": "ProtoGPT (custom decoder-only transformer \u2014 NOT transformers-compatible)",
	"parameter_count": 1013024256,
	"tie_word_embeddings": true,
	"tied_aliases_dropped": [
	"head.weight"
	],
	"checkpoint_step": 373500,
	"torch_dtype": "bfloat16",
	"loader": "model.py::load_base() \u2014 see USAGE.md",
	"root": "/root/stage07_shards_tessera64k_4k",
	"out_dir": "/root/runs/20260615T061150_1b_GATED_h100",
	"device": "cuda",
	"steps": 376000,
	"seq_len": 4096,
	"micro_batch": 4,
	"grad_accum": 4,
	"d_model": 1536,
	"layers": 32,
	"heads": 16,
	"dropout": 0.0,
	"vocab_size": 65536,
	"lr": 0.0002,
	"min_lr": 1e-05,
	"warmup_steps": 200,
	"weight_decay": 0.1,
	"optimizer": "adamw",
	"eval_every": 100,
	"eval_batches": 64,
	"checkpoint_every": 500,
	"log_every": 10,
	"seed": 20260614,
	"param_dtype": "bfloat16",
	"autocast_dtype": "bfloat16",
	"min_free_gb": 8.0,
	"max_runtime_min": 0.0,
	"resume": "/root/checkpoint_resume.pt",
	"save_optimizer": true,
	"no_final_checkpoint": false,
	"no_grad_checkpointing": true,
	"reset_step": true,
	"gated_epoch": true,
	"gated_trust_cursor": true
	}