LiquidAI · fine-tune of LiquidAI/LFM2.5-2.6B-Base
LFM2.5-2.6B
Runs on your machine
Choose a machine
We then show the largest format that fits, from real file sizes where we have them.
Scores
No independent scores for this model in Epoch AI's data. A score is only shown here when Epoch's name for a model matches this repo with certainty; an uncertain match is left out rather than guessed.
Claimed by the maker Claimed
The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.
Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.
Run it yourself
Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 0.13 GB, worked out from the model's config (LiquidAI/LFM2.5-2.6B), at 16-bit precision. Choose a machine above to see a verdict for each format. Method
weights context cache runtime
GGUF for llama.cpp, Ollama, LM Studio and others
LiquidAI/LFM2.5-2.6B-GGUF (the maker's own conversion), 92k downloads. Sizes are the real files on the Hub, all parts added up. Showing the common levels.
| Format | Size | Memory needed | Memory | Get it |
|---|---|---|---|---|
| Q4_K_M | 1.6 GBfile size | 2.2 GB1.6 + 0.13 + 0.50 | Commandsllama.cpp llama-server -hf LiquidAI/LFM2.5-2.6B-GGUF:Q4_K_MOllama ollama run hf.co/LiquidAI/LFM2.5-2.6B-GGUF:Q4_K_MDownload hf download LiquidAI/LFM2.5-2.6B-GGUF --include "LFM2.5-2.6B-Q4_K_M.gguf" --local-dir LFM2.5-2.6B-GGUF | |
| Q5_K_M | 1.8 GBfile size | 2.4 GB1.8 + 0.13 + 0.50 | Commandsllama.cpp llama-server -hf LiquidAI/LFM2.5-2.6B-GGUF:Q5_K_MOllama ollama run hf.co/LiquidAI/LFM2.5-2.6B-GGUF:Q5_K_MDownload hf download LiquidAI/LFM2.5-2.6B-GGUF --include "LFM2.5-2.6B-Q5_K_M.gguf" --local-dir LFM2.5-2.6B-GGUF | |
| Q6_K | 2.1 GBfile size | 2.7 GB2.1 + 0.13 + 0.50 | Commandsllama.cpp llama-server -hf LiquidAI/LFM2.5-2.6B-GGUF:Q6_KOllama ollama run hf.co/LiquidAI/LFM2.5-2.6B-GGUF:Q6_KDownload hf download LiquidAI/LFM2.5-2.6B-GGUF --include "LFM2.5-2.6B-Q6_K.gguf" --local-dir LFM2.5-2.6B-GGUF | |
| Q8_0 | 2.7 GBfile size | 3.3 GB2.7 + 0.13 + 0.50 | Commandsllama.cpp llama-server -hf LiquidAI/LFM2.5-2.6B-GGUF:Q8_0Ollama ollama run hf.co/LiquidAI/LFM2.5-2.6B-GGUF:Q8_0Download hf download LiquidAI/LFM2.5-2.6B-GGUF --include "LFM2.5-2.6B-Q8_0.gguf" --local-dir LFM2.5-2.6B-GGUF | |
| BF16 | 5.0 GBfile size | 5.7 GB5.0 + 0.13 + 0.50 | Commandsllama.cpp llama-server -hf LiquidAI/LFM2.5-2.6B-GGUF:BF16Ollama ollama run hf.co/LiquidAI/LFM2.5-2.6B-GGUF:BF16Download hf download LiquidAI/LFM2.5-2.6B-GGUF --include "LFM2.5-2.6B-BF16.gguf" --local-dir LFM2.5-2.6B-GGUF |
MLX for Apple silicon
| Format | Size | Memory needed | Memory | Repo |
|---|---|---|---|---|
| MLX 4bitLiquidAI/LFM2.5-2.6B-MLX-4bit | 1.5 GBfile size | 2.1 GB1.5 + 0.13 + 0.50 | Download page | |
| MLX 8bitLiquidAI/LFM2.5-2.6B-MLX-8bit | 2.7 GBfile size | 3.3 GB2.7 + 0.13 + 0.50 | Download page |
Original weights size from the safetensors header, for vLLM, SGLang, Transformers
| Format | Size | Memory needed | Memory | Repo |
|---|---|---|---|---|
| BF16original weights | 5.0 GBfile size | 5.6 GB5.0 + 0.13 + 0.50 |
hf download LiquidAI/LFM2.5-2.6B --local-dir LFM2.5-2.6BOther conversions (9; sizes not read)
- LiquidAI/LFM2.5-2.6B-ONNX Other, by the maker, 1.3k downloads
- Arki05/LFM2.5-2.6B-GGUF GGUF, 18k downloads
- litert-community/LFM2.5-2.6B Other, 1.3k downloads
- trymirai/LFM2.5-2.6B-L Other, 471 downloads
- trymirai/LFM2.5-2.6B-M Other, 433 downloads
- plavno/LFM2.5-2.6B-AutoRound-W4A16 GPTQ, w4a16, 81 downloads
- plavno/LFM2.5-2.6B-AutoRound-W8A16 GPTQ, 64 downloads
- vrfai/LFM2.5-2.6B-FP8 FP8, 53 downloads
- kingjones777/LFM2.5-2.6B-NVFP4 NVFP4, 34 downloads
The Hub tags 100 or more quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.
Run it by API
No provider on Hugging Face Inference Providers serves this model (as of 19 Nov 2026, 13:58 UTC). Other hosts may; we only list what the router reports.
Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.
Fine-tunes of this model
- LiquidAI/LFM2.5-2.6B-ONNX1.3kUnreviewed
- LiquidAI/LFM2.5-2.6B-DSpark1.1kUnreviewed
- LiquidAI/LFM2.5-2.6B-MLX-mxfp8264Unreviewed
Planned for this page Not built yet
- Try it in private chat
- A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
- Fine-tune it
- Fine-tuning a few base models from this page and publishing the result here.
- Creators and earnings
- Pages for people who publish fine-tunes and adapters of this model, and what they earn.
- Fund and bounties
- Bounties for new open models, paid from a public fund. How the fund will work.
- Contribute compute
- How to give compute to open model work, with Akash.
Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.