LFM2.5-24B-A2B has 23.972 billion total parameters. The selected Q4_K_M GGUF file is 14.526 GB.
Data snapshot: . 1 published test in this catalog.
Model file and details
Provider
Liquid AI
Total parameters
23,971,587,712
Exact file size
14,526,258,464 bytes
Quantization / precision
Q4_K_M
Weight repository
LiquidAI/LFM2.5-24B-A2B-GGUF
Original model repository
LiquidAI/LFM2.5-24B-A2B
Weight revision
0912d33141f77e10705345a6732bfd83f0252fdc
Model revision
1769516ac588c4c558089c94f27d96b0dc67b183
Runtime notes
Confirm model architecture and quantization support in your runner.
Recorded release date
2026-10-13
Date evidence
Liquid AI announced the 24B-total, 2.45B-active text reasoning model on 13 October 2026. Its official same-publisher GGUF repository supplies the exact 14,526,258,464-byte Q4_K_M artifact, llama.cpp support and a 131,072-token source context limit. The selected quantization has publisher device measurements but no separately reported quality benchmark.
These scores were compiled from the linked sources. These published scores were not independently tested here. Different prompts, precision, thinking budgets and tools can change results.
BFCL v4
65.01 / 100
Choosing and calling functions. Published BFCL v4 scores; tool environments and settings differ.
Benchmark efficiency: 4.475 published score points per GB of selected weights. This ratio is not an inference-speed measure; check the test settings to see whether the score covers the selected quantization.
Test settings: Liquid AI publisher evaluation of the BF16 source checkpoint with vLLM, temperature 0.001 and a 4,096-token maximum output. Hardware, evaluator revision, prompt template and repeated-run uncertainty were not stated.
Publisher result for the BF16 source checkpoint, not an independent test or a quality measurement of this selected Q4_K_M file. Liquid AI separately measured this exact Q4_K_M quantization with Pipette and llama.cpp at a 4K-token input context: 184 output tokens/s on an Apple M5 Max and 114 output tokens/s on an AMD Ryzen AI Max+ 395. Those device results do not disclose RAM configuration, OS, llama.cpp revision, output length, batch or sampling settings, so they do not establish speed or compatibility for another system. The 14,526,258,464-byte weight file excludes runtime and KV-cache headroom. The 131,072-token source limit does not mean every device can run that context. The LFM Open License v1.0 conditions commercial rights on annual revenue below USD 10 million; larger organizations need a separate license. The repository also contains a distinct QAD Q4_0 artifact that is not the selected file and is not represented by this score.
Context and version evidence: The repository and file name identify the selected model. Revision hashes are shown above when recorded. Tested context length, full hardware configuration and run date are not consistently recorded in this snapshot. Only settings explicitly stated in each benchmark record are known.
The 14.526 GB figure covers model weights in this file. It excludes the context cache, runtime, operating system and any separate multimodal files. It is not a measured RAM requirement or a token-generation speed.
These complete configurations pass the initial weight-size check for the selected 14.53 GB file. Allow extra memory for context and software. Check the model’s runtime notes and your inference software before buying.
Listing evidence: 2026-09-25. Amazon.com base price visible in an owner-supplied screenshot reviewed 25 Sep 2026; capture time not supplied. Dated reference, not a live quote. Excludes card signup offers, coupons, tax and shipping.
Listing evidence: 2026-10-03. Apple US launch price announced 22 September 2026: USD 1,699. Dated manufacturer reference, not an Amazon offer; tax and shipping may differ.
Listing evidence: 2026-09-24. GMKtec, EVO-X2 / 64GB / 1TB, recorded 22 Sep 2026; exact ASIN capacity identified 24 Sep. This is a dated reference, not a current Amazon price; tax, shipping and bundle contents may differ.
Memory matches are estimates, not tested compatibility or speed. As an Amazon Associate I earn from qualifying purchases.
Confirm model architecture and quantization support in your runner. The selected weights use 14.526 GB; this excludes context cache, runtime buffers and other programs.
These are the first listed tiers that pass our reserve policy, not measured minimum requirements. System RAM and VRAM remain separate. Use the memory tool to change the reserve for your context and runner.