π Model Description
license: apache-2.0 language:
- en
- tr
- moe
- mixture-of-experts
- finance
- gguf
- quantized
FUMEA-F GGUF β Quantized MoE Model
GGUF quantizations of uaytug/fumea-f,
a 4-expert Mixture-of-Experts model for financial and marketing intelligence.
Quantizations
| File | Quant | Size | Use Case |
|---|---|---|---|
| fumea-f-F16.gguf | F16 | 45.64 GB | GPU inference, full precision |
| fumea-f-BF16.gguf | BF16 | 45.64 GB | GPU inference, bfloat16 |
| fumea-f-Q80.gguf | Q80 | 24.25 GB | Near-lossless |
| fumea-f-Q6K.gguf | Q6K | 18.72 GB | Excellent quality |
| fumea-f-Q5KM.gguf | Q5K_M | 16.23 GB | Very good, balanced |
| fumea-f-Q5KS.gguf | Q5K_S | 15.77 GB | Very good, smaller |
| fumea-f-Q50.gguf | Q50 | 15.77 GB | Good quality |
| fumea-f-Q4KM.gguf | Q4K_M | 13.88 GB | Recommended β best quality/size balance |
| fumea-f-Q4KS.gguf | Q4K_S | 13.09 GB | Good quality, smaller |
| fumea-f-Q40.gguf | Q40 | 12.99 GB | Decent, broad compatibility |
| fumea-f-Q3KL.gguf | Q3K_L | 11.96 GB | Acceptable |
| fumea-f-Q3KM.gguf | Q3K_M | 11.08 GB | Lower quality |
| fumea-f-Q3KS.gguf | Q3K_S | 10.04 GB | Low, tiny |
| fumea-f-Q2K.gguf | Q2K | 8.56 GB | Minimal quality |
| fumea-f-IQ4XS.gguf | IQ4XS | 12.44 GB | imatrix 4-bit |
| fumea-f-IQ4NL.gguf | IQ4NL | 13.10 GB | Non-linear 4-bit |
| fumea-f-IQ3M.gguf | IQ3M | 10.23 GB | imatrix 3-bit M |
Ollama
wget https://huggingface.co/uaytug/fumea-f-gguf/resolve/main/fumea-f-Q4KM.gguf
cat > Modelfile << 'EOF'
FROM ./fumea-f-Q4KM.gguf
PARAMETER temperature 0.6
PARAMETER top_p 0.9
PARAMETER num_ctx 32768
SYSTEM You are FUMEA-F, a financial analysis and marketing intelligence assistant.
EOF
ollama create fumea-f -f Modelfile
ollama run fumea-f
LM Studio
Download any GGUF file, import in LM Studio, and start chatting.
Related
- uaytug/fumea-f β Full MoE model
- uaytug/fumea-f-dense β 8B Dense variant
- uaytug/fumea-f-dense-gguf β Dense GGUF quants
License
Apache 2.0
π GGUF File List
| π Filename | π¦ Size | β‘ Download |
|---|---|---|
|
fumea-f-BF16.gguf
LFS
FP16
|
45.64 GB | Download |
|
fumea-f-F16.gguf
LFS
FP16
|
45.64 GB | Download |
|
fumea-f-IQ3_M.gguf
LFS
Q3
|
10.23 GB | Download |
|
fumea-f-IQ4_NL.gguf
LFS
Q4
|
13.1 GB | Download |
|
fumea-f-IQ4_XS.gguf
LFS
Q4
|
12.44 GB | Download |
|
fumea-f-Q2_K.gguf
LFS
Q2
|
8.56 GB | Download |
|
fumea-f-Q3_K_L.gguf
LFS
Q3
|
11.96 GB | Download |
|
fumea-f-Q3_K_M.gguf
LFS
Q3
|
11.08 GB | Download |
|
fumea-f-Q3_K_S.gguf
LFS
Q3
|
10.04 GB | Download |
|
fumea-f-Q4_0.gguf
Recommended
LFS
Q4
|
12.99 GB | Download |
|
fumea-f-Q4_K_M.gguf
LFS
Q4
|
13.88 GB | Download |
|
fumea-f-Q4_K_S.gguf
LFS
Q4
|
13.09 GB | Download |
|
fumea-f-Q5_0.gguf
LFS
Q5
|
15.77 GB | Download |
|
fumea-f-Q5_K_M.gguf
LFS
Q5
|
16.23 GB | Download |
|
fumea-f-Q5_K_S.gguf
LFS
Q5
|
15.77 GB | Download |
|
fumea-f-Q6_K.gguf
LFS
Q6
|
18.72 GB | Download |
|
fumea-f-Q8_0.gguf
LFS
Q8
|
24.25 GB | Download |