Downloads · 30 days
111
3% of all-time downloads
RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf
anthracite-org_-_magnum-v2-12b-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
111
3% of all-time downloads
All-time downloads
4.2K
Public
Repo size
141 GB
Likes
0
Public
Click a slice to open those files.
.gguf163 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
magnum-v2-12b - GGUF
| Name | Quant method | Size |
|---|---|---|
| magnum-v2-12b.Q2_K.gguf | Q2_K | 4.46GB |
| magnum-v2-12b.IQ3_XS.gguf | IQ3_XS | 4.94GB |
| magnum-v2-12b.IQ3_S.gguf | IQ3_S | 5.18GB |
| magnum-v2-12b.Q3_K_S.gguf | Q3_K_S | 5.15GB |
| magnum-v2-12b.IQ3_M.gguf | IQ3_M | 5.33GB |
| magnum-v2-12b.Q3_K.gguf | Q3_K | 5.67GB |
| magnum-v2-12b.Q3_K_M.gguf | Q3_K_M | 5.67GB |
| magnum-v2-12b.Q3_K_L.gguf | Q3_K_L | 6.11GB |
| magnum-v2-12b.IQ4_XS.gguf | IQ4_XS | 6.33GB |
| magnum-v2-12b.Q4_0.gguf | Q4_0 | 6.59GB |
| magnum-v2-12b.IQ4_NL.gguf | IQ4_NL | 6.65GB |
| magnum-v2-12b.Q4_K_S.gguf | Q4_K_S | 6.63GB |
| magnum-v2-12b.Q4_K.gguf | Q4_K | 6.96GB |
| magnum-v2-12b.Q4_K_M.gguf | Q4_K_M | 6.96GB |
| magnum-v2-12b.Q4_1.gguf | Q4_1 | 7.26GB |
| magnum-v2-12b.Q5_0.gguf | Q5_0 | 7.93GB |
| magnum-v2-12b.Q5_K_S.gguf | Q5_K_S | 7.93GB |
| magnum-v2-12b.Q5_K.gguf | Q5_K | 8.13GB |
| magnum-v2-12b.Q5_K_M.gguf | Q5_K_M | 8.13GB |
| magnum-v2-12b.Q5_1.gguf | Q5_1 | 8.61GB |
| magnum-v2-12b.Q6_K.gguf | Q6_K | 9.37GB |
| magnum-v2-12b.Q8_0.gguf | Q8_0 | 12.13GB |
language:

This is the fourth in a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet and Opus. This model is fine-tuned on top of Mistral-Nemo-Base-2407.
Model has been Instruct tuned with the ChatML formatting. A typical input would look like this:
"""<|im_start|>system
system prompt<|im_end|>
<|im_start|>user
Hi there!<|im_end|>
<|im_start|>assistant
Nice to meet you!<|im_end|>
<|im_start|>user
Can I ask a question?<|im_end|>
<|im_start|>assistant
"""
This model has been a team effort, and the credits goes to all members of Anthracite.
The training was done for 2 epochs. We used 8x NVIDIA H100 Tensor Core GPUs for the full-parameter fine-tuning of the model.
...
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 18.68 |
| IFEval (0-Shot) | 37.62 |
| BBH (3-Shot) | 28.79 |
| MATH Lvl 5 (4-Shot) | 4.76 |
| GPQA (0-shot) | 5.48 |
| MuSR (0-shot) | 11.37 |
| MMLU-PRO (5-shot) | 24.08 |