What are the hardware requirements for KOLIBRI 1 DERISKED BF16?

KOLIBRI 1 DERISKED BF16 requires approximately 156.3 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed KOLIBRI 1 DERISKED BF16 and how large is it?

KOLIBRI 1 DERISKED BF16 is developed by Blackfrost-AI, a model with 78.1b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate KOLIBRI 1 DERISKED BF16?

KOLIBRI 1 DERISKED BF16 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

KOLIBRI 1 DERISKED BF16 by Blackfrost-AI — VRAM 156.3GB, 256K context

Name: KOLIBRI 1 DERISKED BF16
Author: Blackfrost-AI

KOLIBRI 1 DERISKED BF16 is an open-source language model by Blackfrost-AI. Features: 78.1b LLM, VRAM: 156.3GB, Context: 256K, License: apache-2.0, LLM Explorer Score: 0.33.

Base model:aleph-alpha/kolibri... Base model:finetune:aleph-alph... Bfloat16 Conversational De En Kolibri1 Long-context Moe Reasoning Region:us Safetensors Sharded Tensorflow Tool-calling Vllm

Model Card on HF 🤗: https://huggingface.co/Blackfrost-AI/KOLIBRI-1-DERISKED-BF16

KOLIBRI 1 DERISKED BF16 Parameters and Internals

LLM Name	KOLIBRI 1 DERISKED BF16
Repository 🤗	https://huggingface.co/Blackfrost-AI/KOLIBRI-1-DERISKED-BF16
Base Model(s)	Kolibri 1 BF16 Aleph-Alpha/Kolibri-1-BF16
Model Size	78.1b
Required VRAM	156.3 GB
Updated	2026-10-10
Maintainer	Blackfrost-AI
Model Type	kolibri1
Model Files	5.0 GB: 1-of-32 5.0 GB: 2-of-32 5.0 GB: 3-of-32 5.0 GB: 4-of-32 5.0 GB: 5-of-32 5.0 GB: 6-of-32 5.0 GB: 7-of-32 5.0 GB: 8-of-32 5.0 GB: 9-of-32 5.0 GB: 10-of-32 5.0 GB: 11-of-32 5.0 GB: 12-of-32 5.0 GB: 13-of-32 5.0 GB: 14-of-32 5.0 GB: 15-of-32 5.0 GB: 16-of-32 5.0 GB: 17-of-32 5.0 GB: 18-of-32 5.0 GB: 19-of-32 5.0 GB: 20-of-32 5.0 GB: 21-of-32 5.0 GB: 22-of-32 5.0 GB: 23-of-32 5.0 GB: 24-of-32 5.0 GB: 25-of-32 5.0 GB: 26-of-32 5.0 GB: 27-of-32 5.0 GB: 28-of-32 5.0 GB: 29-of-32 5.0 GB: 30-of-32 5.0 GB: 31-of-32 1.3 GB: 32-of-32
Supported Languages	de en
Model Architecture	Kolibri1ForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<\|endoftext\|>
Vocabulary Size	128000
Errors	replace

Rank the KOLIBRI 1 DERISKED BF16 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to KOLIBRI 1 DERISKED BF16

Best Alternatives	Context / RAM	Downloads	Likes
Kolibri 1	256K / 78.8 GB	8474	845
Kolibri 1 BF16	256K / 156.3 GB	1449	56
Kolibri 1 Mlx Mixed 4 8 Bit	256K / 44.9 GB	1341	5
Kolibri 1 Heretic	256K / 78.8 GB	69	2
Kolibri 1	256K / 78.8 GB	107	1
KOLIBRI 1 DERISKED FP8	256K / 78.8 GB	46	1
Kolibri 1 MLX 3bit	256K / 35.1 GB	1126	3
Kolibri 1 MLX 4bit	256K / 44.1 GB	955	3
Kolibri 1 MLX 6bit	256K / 63.1 GB	944	2
Kolibri 1 Mlx 4bit	256K / 43.7 GB	615	2

Note: green Score (e.g. "73.2") means that the model is better than Blackfrost-AI/KOLIBRI-1-DERISKED-BF16.