Update README.md
Browse files
README.md
CHANGED
|
@@ -21,7 +21,7 @@ language:
|
|
| 21 |
**Aramis-2B-BitNet** *(2.41B params / Context Length: Maximum sequence length of 4096 tokens)*
|
| 22 |
A compact, agent-oriented small language model focused on contextual reasoning, language understanding and multi-turn instruction following.
|
| 23 |
Built with an iterative post-training recipe: bilingual DPO (FR+EN) + model merging of FR-centric and EN-centric variants.
|
| 24 |
-
Runs natively as BitNet 1.58-bit (ternary) and is available in GGUF 1.58-bit, lossless
|
| 25 |
|
| 26 |
**Why BitNet (and why this model)**
|
| 27 |
- BitNet b1.58 uses ternary weights (−1,0,+1) with abs-mean scaling : very low memory & energy, great CPU/edge throughput, unlike classic FP/INT SLMs. For more details on the underlying architecture and efficiency of BitNet, please refer to the official Microsoft Research publication: [BitNet b1.58 2B4T Technical Report](https://arxiv.org/abs/2504.12285)
|
|
|
|
| 21 |
**Aramis-2B-BitNet** *(2.41B params / Context Length: Maximum sequence length of 4096 tokens)*
|
| 22 |
A compact, agent-oriented small language model focused on contextual reasoning, language understanding and multi-turn instruction following.
|
| 23 |
Built with an iterative post-training recipe: bilingual DPO (FR+EN) + model merging of FR-centric and EN-centric variants.
|
| 24 |
+
Runs natively as BitNet 1.58-bit (ternary) and is available in GGUF 1.58-bit, lossless from the BF16 checkpoint.
|
| 25 |
|
| 26 |
**Why BitNet (and why this model)**
|
| 27 |
- BitNet b1.58 uses ternary weights (−1,0,+1) with abs-mean scaling : very low memory & energy, great CPU/edge throughput, unlike classic FP/INT SLMs. For more details on the underlying architecture and efficiency of BitNet, please refer to the official Microsoft Research publication: [BitNet b1.58 2B4T Technical Report](https://arxiv.org/abs/2504.12285)
|