Total runs:
348.3K
Run Growth:
267.3K
Growth Rate:
76.17%
Sponsored by
APIMart
AtomicChat / lfm25-8b-a1b-GGUF
Model's Last Updated: July 23 2026
text-generation
Introduction of lfm25-8b-a1b-GGUF
Model Details of lfm25-8b-a1b-GGUF
LFM2.5 8B A1B , self-quantized to GGUF by Atomic Chat . Built straight from Liquid AI's original weights with a per-tensor importance matrix. Runs fully offline.
Highlights
- Sparse MoE : 8.3B total parameters, only 1.5B active per token.
- LFM2 hybrid architecture : 24 layers (18 double-gated LIV convolution blocks + 6 GQA attention), built on LFM2 with extended pre-training and reinforcement learning.
- On-device assistant : designed to chain tool calls and follow complex instructions, with day-one support for llama.cpp, MLX, vLLM and SGLang.
- Reasoning model : assistant turns include an explicit chain of thought before the final answer.
- 128K context , 128,000 vocabulary, trained on a 38 trillion token budget.
- Multilingual : English, Arabic, Chinese, French, German, Italian, Japanese, Korean, Portuguese, Spanish.
These GGUFs are self-quantized from the original weights , not a repack. The importance matrix keeps low-bit quants closer to the full-precision model.
Always pass
--jinjaso the LFM2.5 8B A1B chat template is applied. Without it the model can emit malformed turns.
Model Overview
| Property | Value |
|---|---|
| Base model |
LiquidAI/LFM2.5-8B-A1B
|
| Total / active parameters | 8.3B total, 1.5B active (MoE) |
| Layers | 24 (18 LIV conv + 6 GQA) |
| Context length | 128,000 |
| Architecture | LFM2.5 hybrid (built on LFM2, extended pre-training + RL) |
| This repo | GGUF quants (imatrix) |
Scores are Liquid AI's published results for the base
LiquidAI/LFM2.5-8B-A1B
. Quantization preserves the large majority of this;
Q4_K_M
and up sit within a point or two of full precision.
Choosing a quant
| Quant | Size | Notes |
|---|---|---|
Q2_K
|
3.2 GB | Smallest. Minimal RAM, clear quality drop. |
IQ3_M
|
3.8 GB | Beats Q3 at similar size thanks to imatrix. Best low-RAM pick. |
Q3_K_M
|
4.1 GB | Low quality but usable. |
Q3_K_L
|
4.4 GB | A step above Q3_K_M. |
IQ4_XS
|
4.6 GB | Excellent quality for size. Recommended low-bit. |
Q4_K_S
|
4.9 GB | Compact Q4, fast. |
Q4_K_M
|
5.2 GB | Recommended default. Best balance of size, speed and quality. |
UD-Q4_K_XL
|
5.2 GB | Dynamic. Embeddings and output kept at Q8_0 for higher quality at a Q4 footprint. |
Q5_K_S
|
5.9 GB | Higher quality. |
Q5_K_M
|
6.0 GB | Higher quality, low loss. |
Q6_K
|
7.0 GB | Near lossless. |
Q8_0
|
9.0 GB | Effectively lossless, reference quality. |
Pick the largest file that fits your (V)RAM with room for context.
Q4_K_MorUD-Q4_K_XLis the sweet spot for most setups;Q6_KorQ8_0for maximum fidelity.
Get started
Run LFM2.5 8B A1B locally with:
-
Atomic Chat
:
the easiest path. Open the app, search
AtomicChat/lfm25-8b-a1b-GGUF, pick a quant, hit Use this model . -
llama.cpp:
llama-server -hf AtomicChat/lfm25-8b-a1b-GGUF:Q4_K_M --jinja -c 8192 -
Ollama:
ollama run hf.co/AtomicChat/lfm25-8b-a1b-GGUF:Q4_K_M - LM Studio / Jan: search the repo id, download any quant.
Best practices
| Parameter | Value |
|---|---|
| temperature | 0.2 |
| top_k | 80 |
| repetition_penalty | 1.05 |
Liquid AI's recommended generation parameters.
Run in llama.cpp
git clone https://github.com/ggerganov/llama.cpp
cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=ON
cmake --build llama.cpp/build --config Release -j --target llama-cli llama-server
./llama.cpp/build/bin/llama-server \
-hf AtomicChat/lfm25-8b-a1b-GGUF:UD-Q4_K_XL \
--jinja -ngl 99 -c 8192 -fa on
How these were made
-
Download
LiquidAI/LFM2.5-8B-A1B(original weights). - Convert to f16 GGUF with llama.cpp .
-
Build an importance matrix over
calibration_datav3(100 chunks). -
Quantize the full ladder with
--imatrix. -
UD-Q4_K_XLadditionally pins the token-embedding and output tensors toQ8_0.
License
Released by Liquid AI under their LFM1.0 license. Quantized by Atomic Chat.
Runs of AtomicChat lfm25-8b-a1b-GGUF on huggingface.co
429
Total runs
0
24-hour runs
63
3-day runs
-399
7-day runs
-426
30-day runs
More Information About lfm25-8b-a1b-GGUF huggingface.co Model
Url of lfm25-8b-a1b-GGUF
lfm25-8b-a1b-GGUF huggingface.co Url
Provider of lfm25-8b-a1b-GGUF huggingface.co
Other API from AtomicChat
Updated:August 09 2026
Total runs:
257.0K
Run Growth:
226.1K
Growth Rate:
100.00%
Total runs:
82.9K
Run Growth:
-88.1K
Growth Rate:
-74.16%
Total runs:
77.4K
Run Growth:
81.9K
Growth Rate:
89.97%
Total runs:
73.4K
Run Growth:
41.5K
Growth Rate:
56.54%
Total runs:
73.1K
Run Growth:
40.5K
Growth Rate:
55.40%
Total runs:
73.0K
Run Growth:
40.3K
Growth Rate:
55.22%
Total runs:
54.8K
Run Growth:
58.7K
Growth Rate:
95.03%
Total runs:
6.9K
Run Growth:
1.6K
Growth Rate:
23.81%
Total runs:
6.2K
Run Growth:
383
Growth Rate:
6.18%
Total runs:
4.7K
Run Growth:
-3.1K
Growth Rate:
-66.17%
Total runs:
4.2K
Run Growth:
-13.7K
Growth Rate:
-323.94%
Total runs:
4.1K
Run Growth:
-8.3K
Growth Rate:
-200.53%
Total runs:
4.0K
Run Growth:
-1.9K
Growth Rate:
-46.89%
Total runs:
3.5K
Run Growth:
-11.7K
Growth Rate:
-333.71%
Total runs:
3.5K
Run Growth:
-1.8K
Growth Rate:
-52.85%
Total runs:
3.4K
Run Growth:
550
Growth Rate:
16.18%
Total runs:
3.1K
Run Growth:
-18.2K
Growth Rate:
-583.49%
Total runs:
3.1K
Run Growth:
739
Growth Rate:
23.69%
Total runs:
2.6K
Run Growth:
-8.3K
Growth Rate:
-314.46%
Total runs:
2.5K
Run Growth:
-1.5K
Growth Rate:
-57.65%
Total runs:
2.2K
Run Growth:
564
Growth Rate:
25.37%
Total runs:
2.0K
Run Growth:
1.8K
Growth Rate:
100.00%
Total runs:
2.0K
Run Growth:
-14.2K
Growth Rate:
-718.94%
Total runs:
2.0K
Run Growth:
771
Growth Rate:
39.42%
Total runs:
1.9K
Run Growth:
-14.7K
Growth Rate:
-782.90%
Total runs:
1.7K
Run Growth:
657
Growth Rate:
37.65%
Total runs:
1.7K
Run Growth:
-361
Growth Rate:
-20.89%
Total runs:
1.6K
Run Growth:
1.6K
Growth Rate:
100.00%
Total runs:
1.5K
Run Growth:
699
Growth Rate:
45.36%
Total runs:
1.2K
Run Growth:
1.2K
Growth Rate:
100.00%
Total runs:
1.2K
Run Growth:
-2.0K
Growth Rate:
-164.53%
Total runs:
1.2K
Run Growth:
-75
Growth Rate:
-6.17%
Total runs:
1.1K
Run Growth:
-907
Growth Rate:
-79.08%
Total runs:
1.1K
Run Growth:
1.1K
Growth Rate:
100.00%
Total runs:
1.1K
Run Growth:
1.1K
Growth Rate:
100.00%
Total runs:
1.0K
Run Growth:
1.0K
Growth Rate:
100.00%
Total runs:
887
Run Growth:
220
Growth Rate:
24.80%
Total runs:
753
Run Growth:
-27
Growth Rate:
-3.59%
Total runs:
688
Run Growth:
-3.5K
Growth Rate:
-506.10%
Total runs:
668
Run Growth:
668
Growth Rate:
100.00%
Total runs:
660
Run Growth:
-7.3K
Growth Rate:
-1115.50%
Total runs:
640
Run Growth:
-3.3K
Growth Rate:
-523.70%
Total runs:
602
Run Growth:
602
Growth Rate:
100.00%
Total runs:
599
Run Growth:
599
Growth Rate:
100.00%
Total runs:
580
Run Growth:
253
Growth Rate:
43.62%
Total runs:
565
Run Growth:
100
Growth Rate:
17.70%
Total runs:
550
Run Growth:
-1.2K
Growth Rate:
-209.27%
Total runs:
536
Run Growth:
-142
Growth Rate:
-26.49%
Total runs:
466
Run Growth:
-1.2K
Growth Rate:
-251.72%
Total runs:
451
Run Growth:
-590
Growth Rate:
-130.82%
Total runs:
424
Run Growth:
424
Growth Rate:
100.00%
Total runs:
398
Run Growth:
-174
Growth Rate:
-43.72%
Total runs:
392
Run Growth:
306
Growth Rate:
78.06%
Total runs:
390
Run Growth:
-1.4K
Growth Rate:
-354.87%
Total runs:
361
Run Growth:
296
Growth Rate:
81.99%
Total runs:
323
Run Growth:
-279
Growth Rate:
-86.38%
Total runs:
306
Run Growth:
306
Growth Rate:
100.00%
Total runs:
300
Run Growth:
-637
Growth Rate:
-212.33%
Total runs:
298
Run Growth:
298
Growth Rate:
100.00%
Total runs:
296
Run Growth:
296
Growth Rate:
100.00%
Total runs:
283
Run Growth:
-148
Growth Rate:
-52.30%
Total runs:
279
Run Growth:
-1.5K
Growth Rate:
-529.39%
Total runs:
245
Run Growth:
-378
Growth Rate:
-154.29%
Total runs:
237
Run Growth:
237
Growth Rate:
100.00%
Total runs:
231
Run Growth:
-171
Growth Rate:
-74.03%
Total runs:
223
Run Growth:
-861
Growth Rate:
-386.10%
Total runs:
215
Run Growth:
-74
Growth Rate:
-34.42%
Total runs:
214
Run Growth:
-1.5K
Growth Rate:
-692.99%
Total runs:
211
Run Growth:
211
Growth Rate:
100.00%
Total runs:
195
Run Growth:
-6.5K
Growth Rate:
-3312.82%
Total runs:
189
Run Growth:
-105
Growth Rate:
-55.56%
Total runs:
151
Run Growth:
-877
Growth Rate:
-580.79%
Total runs:
150
Run Growth:
-622
Growth Rate:
-414.67%
Total runs:
149
Run Growth:
-68
Growth Rate:
-45.64%
Total runs:
133
Run Growth:
-619
Growth Rate:
-465.41%
Total runs:
133
Run Growth:
-242
Growth Rate:
-181.95%
Total runs:
130
Run Growth:
-344
Growth Rate:
-264.62%
Total runs:
119
Run Growth:
-376
Growth Rate:
-315.97%
Total runs:
117
Run Growth:
117
Growth Rate:
100.00%
Total runs:
117
Run Growth:
104
Growth Rate:
88.89%
Total runs:
115
Run Growth:
-898
Growth Rate:
-780.87%
Total runs:
106
Run Growth:
-761
Growth Rate:
-717.92%
Total runs:
104
Run Growth:
-805
Growth Rate:
-774.04%
Total runs:
99
Run Growth:
-918
Growth Rate:
-927.27%
Total runs:
98
Run Growth:
-1.0K
Growth Rate:
-1060.20%
Total runs:
97
Run Growth:
-200
Growth Rate:
-206.19%
Total runs:
96
Run Growth:
-331
Growth Rate:
-344.79%
Total runs:
94
Run Growth:
82
Growth Rate:
87.23%
Total runs:
93
Run Growth:
-1.2K
Growth Rate:
-1291.40%
Total runs:
78
Run Growth:
-975
Growth Rate:
-1250.00%
Total runs:
76
Run Growth:
-362
Growth Rate:
-476.32%
Total runs:
65
Run Growth:
-14
Growth Rate:
-21.54%
Total runs:
62
Run Growth:
-31
Growth Rate:
-50.00%
Total runs:
55
Run Growth:
-35
Growth Rate:
-63.64%
Total runs:
48
Run Growth:
0
Growth Rate:
0.00%
Total runs:
23
Run Growth:
-37
Growth Rate:
-160.87%