Total runs:
312.7K
Run Growth:
277.1K
Growth Rate:
88.62%
Sponsored by
YouArt
AtomicChat / LFM2.5-8B-A1B-GGUF
Model's Last Updated: July 23 2026
text-generation
Introduction of LFM2.5-8B-A1B-GGUF
Model Details of LFM2.5-8B-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, so this is not a repack of somebody else's files. Runs fully offline.
Highlights
- 8.5B parameters : the weights this repo quantizes.
- Context length : 128,000 tokens (125K), as published by Liquid AI.
- 24 layers : Mixture-of-Experts.
- Full imatrix ladder : every quant is calibrated with an importance matrix.
- On-device personal assistant : Designed to power real-life applications, chaining tool calls, and following complex instructions on all devices.
- Compressed performance : Competitive with much larger dense and MoE models on instruction following and agentic tasks.
- Unmatched throughput : Fastest in its size class on both CPU and GPU inference, with day-one support for llama.cpp, MLX, vLLM, and SGLang.
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
|
| Parameters | 8.5B |
| Layers | 24 |
| Experts | 32 routed (top-4) |
| Context length | 128,000 tokens (125K) |
| Vocabulary | 128,000 |
| Modalities | Text |
| Architecture |
Mixture-of-Experts, 32 experts (top-4), 32 attention heads over 8 KV heads,
Lfm2MoeForCausalLM
|
| This repo |
GGUF quants (imatrix). Quants:
Q2_K
,
IQ3_M
,
Q3_K_M
,
Q3_K_L
,
IQ4_XS
,
Q4_K_S
,
Q4_K_M
,
UD-Q4_K_XL
,
Q5_K_S
,
Q5_K_M
,
Q6_K
,
Q8_0
|
Scores are Liquid AI's published results for the base
LiquidAI/LFM2.5-8B-A1B
, not our own measurements. Quantization preserves the large majority of this;
Q4_K_M
and up stay close to full precision.
Choosing a quant
| Quant | Size | Notes |
|---|---|---|
Q2_K
|
3.2 GB | Smallest K-quant. Minimal RAM, clear quality drop. |
IQ3_M
|
3.8 GB | Beats Q3 at a 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 4-bit, 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, slightly more compact than Q5_K_M. |
Q5_K_M
|
6.0 GB | Higher quality, low loss. |
Q6_K
|
7.0 GB | Near lossless, noticeably lighter than Q8_0. |
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 sampling configuration for
LiquidAI/LFM2.5-8B-A1B
.
Run in llama.cpp
git clone https://github.com/ggml-org/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:Q4_K_M \
--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 our calibration corpus.
-
Quantize the ladder with
--imatrix. -
UD-Q4_K_XLadditionally pins the token-embedding and output tensors toQ8_0.
License
Original model by Liquid AI, released under the other license. Full terms: other . Quantized by Atomic Chat.
Runs of AtomicChat LFM2.5-8B-A1B-GGUF on huggingface.co
563
Total runs
-11
24-hour runs
24
3-day runs
104
7-day runs
-45
30-day runs
More Information About LFM2.5-8B-A1B-GGUF huggingface.co Model
Url of LFM2.5-8B-A1B-GGUF
LFM2.5-8B-A1B-GGUF huggingface.co Url
Provider of LFM2.5-8B-A1B-GGUF huggingface.co
Other API from AtomicChat
Updated:August 19 2026
Total runs:
307.2K
Run Growth:
237.2K
Growth Rate:
77.24%
Total runs:
88.9K
Run Growth:
79.8K
Growth Rate:
89.74%
Total runs:
73.0K
Run Growth:
53.1K
Growth Rate:
72.78%
Total runs:
72.9K
Run Growth:
56.8K
Growth Rate:
78.00%
Total runs:
72.4K
Run Growth:
54.3K
Growth Rate:
74.94%
Total runs:
59.8K
Run Growth:
59.8K
Growth Rate:
100.00%
Total runs:
59.5K
Run Growth:
56.4K
Growth Rate:
94.84%
Total runs:
7.5K
Run Growth:
-1.3K
Growth Rate:
-16.64%
Total runs:
7.5K
Run Growth:
5.4K
Growth Rate:
71.79%
Total runs:
5.6K
Run Growth:
113
Growth Rate:
2.03%
Total runs:
5.0K
Run Growth:
-14.0K
Growth Rate:
-280.59%
Total runs:
4.6K
Run Growth:
-1.9K
Growth Rate:
-41.02%
Total runs:
4.6K
Run Growth:
-12.8K
Growth Rate:
-279.45%
Total runs:
4.4K
Run Growth:
-9.7K
Growth Rate:
-219.28%
Total runs:
3.3K
Run Growth:
-2.8K
Growth Rate:
-83.75%
Total runs:
2.6K
Run Growth:
436
Growth Rate:
17.07%
Total runs:
2.5K
Run Growth:
-1.3K
Growth Rate:
-53.05%
Total runs:
2.3K
Run Growth:
-11.8K
Growth Rate:
-511.13%
Total runs:
2.3K
Run Growth:
-1.4K
Growth Rate:
-61.26%
Total runs:
2.2K
Run Growth:
-2.8K
Growth Rate:
-126.64%
Total runs:
2.0K
Run Growth:
295
Growth Rate:
14.55%
Total runs:
2.0K
Run Growth:
-9.4K
Growth Rate:
-478.02%
Total runs:
1.8K
Run Growth:
-15.0K
Growth Rate:
-850.62%
Total runs:
1.7K
Run Growth:
657
Growth Rate:
37.65%
Total runs:
1.7K
Run Growth:
960
Growth Rate:
55.81%
Total runs:
1.6K
Run Growth:
616
Growth Rate:
38.02%
Total runs:
1.6K
Run Growth:
1.6K
Growth Rate:
100.00%
Total runs:
1.5K
Run Growth:
-594
Growth Rate:
-39.49%
Total runs:
1.3K
Run Growth:
-2.1K
Growth Rate:
-163.80%
Total runs:
1.2K
Run Growth:
1.2K
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:
976
Run Growth:
976
Growth Rate:
100.00%
Total runs:
964
Run Growth:
964
Growth Rate:
100.00%
Total runs:
934
Run Growth:
-324
Growth Rate:
-34.69%
Total runs:
867
Run Growth:
-945
Growth Rate:
-109.00%
Total runs:
778
Run Growth:
-288
Growth Rate:
-37.02%
Total runs:
654
Run Growth:
-3.6K
Growth Rate:
-547.71%
Total runs:
618
Run Growth:
-7.7K
Growth Rate:
-1249.03%
Total runs:
602
Run Growth:
602
Growth Rate:
100.00%
Total runs:
555
Run Growth:
555
Growth Rate:
100.00%
Total runs:
535
Run Growth:
-3.3K
Growth Rate:
-619.07%
Total runs:
523
Run Growth:
-623
Growth Rate:
-119.12%
Total runs:
523
Run Growth:
24
Growth Rate:
4.59%
Total runs:
480
Run Growth:
-467
Growth Rate:
-97.29%
Total runs:
465
Run Growth:
-1.3K
Growth Rate:
-271.61%
Total runs:
430
Run Growth:
-1.4K
Growth Rate:
-320.70%
Total runs:
429
Run Growth:
-426
Growth Rate:
-99.30%
Total runs:
426
Run Growth:
410
Growth Rate:
96.24%
Total runs:
394
Run Growth:
-542
Growth Rate:
-137.56%
Total runs:
391
Run Growth:
-1.3K
Growth Rate:
-329.41%
Total runs:
379
Run Growth:
379
Growth Rate:
100.00%
Total runs:
357
Run Growth:
-1.4K
Growth Rate:
-405.88%
Total runs:
336
Run Growth:
336
Growth Rate:
100.00%
Total runs:
313
Run Growth:
-37
Growth Rate:
-11.82%
Total runs:
296
Run Growth:
-568
Growth Rate:
-191.89%
Total runs:
283
Run Growth:
-1.4K
Growth Rate:
-496.11%
Total runs:
281
Run Growth:
281
Growth Rate:
100.00%
Total runs:
273
Run Growth:
273
Growth Rate:
100.00%
Total runs:
272
Run Growth:
-141
Growth Rate:
-51.84%
Total runs:
269
Run Growth:
269
Growth Rate:
100.00%
Total runs:
257
Run Growth:
-843
Growth Rate:
-328.02%
Total runs:
245
Run Growth:
-1.1K
Growth Rate:
-433.88%
Total runs:
237
Run Growth:
-5.6K
Growth Rate:
-2350.63%
Total runs:
235
Run Growth:
170
Growth Rate:
72.34%
Total runs:
231
Run Growth:
-794
Growth Rate:
-343.72%
Total runs:
218
Run Growth:
218
Growth Rate:
100.00%
Total runs:
211
Run Growth:
-656
Growth Rate:
-310.90%
Total runs:
201
Run Growth:
-305
Growth Rate:
-151.74%
Total runs:
199
Run Growth:
-20
Growth Rate:
-10.05%
Total runs:
189
Run Growth:
189
Growth Rate:
100.00%
Total runs:
177
Run Growth:
-957
Growth Rate:
-540.68%
Total runs:
169
Run Growth:
-223
Growth Rate:
-131.95%
Total runs:
159
Run Growth:
-774
Growth Rate:
-486.79%
Total runs:
144
Run Growth:
-283
Growth Rate:
-196.53%
Total runs:
141
Run Growth:
-419
Growth Rate:
-297.16%
Total runs:
139
Run Growth:
-707
Growth Rate:
-508.63%
Total runs:
137
Run Growth:
-313
Growth Rate:
-228.47%
Total runs:
137
Run Growth:
-228
Growth Rate:
-166.42%
Total runs:
136
Run Growth:
-818
Growth Rate:
-601.47%
Total runs:
127
Run Growth:
-831
Growth Rate:
-654.33%
Total runs:
122
Run Growth:
-700
Growth Rate:
-573.77%
Total runs:
119
Run Growth:
-744
Growth Rate:
-625.21%
Total runs:
117
Run Growth:
-975
Growth Rate:
-833.33%
Total runs:
117
Run Growth:
117
Growth Rate:
100.00%
Total runs:
117
Run Growth:
104
Growth Rate:
88.89%
Total runs:
116
Run Growth:
-1.1K
Growth Rate:
-964.66%
Total runs:
114
Run Growth:
-894
Growth Rate:
-784.21%
Total runs:
107
Run Growth:
-253
Growth Rate:
-236.45%
Total runs:
106
Run Growth:
-423
Growth Rate:
-399.06%
Total runs:
94
Run Growth:
82
Growth Rate:
87.23%
Total runs:
91
Run Growth:
-440
Growth Rate:
-483.52%
Total runs:
78
Run Growth:
22
Growth Rate:
28.21%
Total runs:
62
Run Growth:
-19
Growth Rate:
-30.65%
Total runs:
55
Run Growth:
-26
Growth Rate:
-47.27%
Total runs:
48
Run Growth:
-6
Growth Rate:
-12.50%
Total runs:
48
Run Growth:
4
Growth Rate:
8.33%