27 Oct 2026
LFM3-1.7B
Small weights can still carry a growing cache
1.73B reported total · text
LFM / Convolution / GQA MoE
Active parameters are a compute clue, not a device budget
New to model sizes or GPU memory? Start with weights, parameters and quantization →
A candidate for private agents when a larger local model is acceptable. Keep its original BF16 checkpoint distinct from the FP8, GGUF and MLX distributions.
Publisher-reported rounded count; the pinned tensor index is a separate payload measurement.
Evaluate it for
Local reasoning and tool workflows with an artifact-specific serving path.
Choose another path when
Memory favors a much smaller dense checkpoint or the application needs image inputs.
Release context
The 40-layer design combines thirty short convolutions with ten grouped-query attention layers. Routing selects four of 64 experts; their stored tensors remain part of residency.
Publisher release post; not the earlier repository creation date.
Deployment starting points
Start with an exact artifact and an engine that supports it. A configuration below is evidence of a documented path; its device count does not promise a particular throughput or concurrency.
Memory favors a much smaller dense checkpoint or the application needs image inputs.
The pinned model card establishes the architecture. It does not, by itself, prove that a particular GPU count runs this checkpoint at your target context. The capacity screen below helps rule out undersized allocations; a successful load and workload test are still required.
This is the index’s reported tensor payload, in decimal GB. It excludes file headers and runtime memory. Mixed precision, conversion and host/device placement determine how much GPU memory the loaded model needs.
Inspect the exact index metadata →This estimates inference memory for a hypothetical uniform precision. It is useful for rejecting an allocation that is too small. It does not establish a working deployment or estimate training memory.
| Representation | Approximate weight floor | What remains to budget |
|---|---|---|
| 16-bit uniform | 48 GB | Cache, modality state, workspace, quantization metadata and uneven sharding |
| 8-bit uniform | 24 GB | Cache, modality state, workspace, quantization metadata and uneven sharding |
| 4-bit uniform | 12 GB | Cache, modality state, workspace, quantization metadata and uneven sharding |
Enter the total across the whole model replica, not per GPU: KV or recurrent cache for all concurrent requests, image/audio state and execution workspace. There is no universal default. We additionally leave 15% of each GPU unused.
Enter a request-and-buffer budget to calculate the aggregate capacity requirement.
A count of one does not prove the engine supports this GPU. A count above one does not prove the model shards evenly or communicates efficiently. Publisher-reported rounded count; the pinned tensor index is a separate payload measurement.
Architecture in practice
Small active compute does not shrink the expert store.
Count per-request state separately from shared weights.
Mechanism schematic based on the pinned model card and configuration. It explains a design principle; it is not a full implementation graph.
| Exact checkpoint | LiquidAI/LFM2.5-24B-A2B |
|---|---|
| Text model type | lfm2_moe |
| Layers | 40 |
| Hidden width | 2,048 |
| Attention / KV heads | 32 / 8 |
| Head dimension | Not verified |
| Routed / selected experts | 64 / 4 |
| Configured positions | 131,072 |
| Documented extension | Not recorded for this checkpoint |
| Inputs → output | text → text |
| License metadata | LFM Open License v1.0 · custom terms |
30 short convolution layers; 10 full attention layers. These counts describe the configured pattern.
Open weights do not imply unrestricted use. Read the applicable license terms linked from the pinned card.
A useful comparison
Use the smaller LFM3 for a constrained-memory comparison and Qwen for a larger hybrid-agent comparison; rerun the same tools and stopping rules.
27 Oct 2026
Small weights can still carry a growing cache
1.73B reported total · text
16 Apr 2026
A smaller expert store for coding-agent experiments
35B reported total · text + image + video
These are editorial comparison candidates. We have not run a matched quality or serving benchmark, so this is not a ranking.
Prepare to explain it
Reconcile the 24B rounded total, 2.45B active path and 47,943,180,288-byte tensor payload before choosing a device.
Attention layouts · Quantization · Courses and worked examples
For a deployment evaluation, record exact weights, precision, engine, device count, interconnect, prompt/output lengths and concurrency. Report task success, errors, TTFT, TPOT and useful throughput together.
Our assessment is editorial judgment based on the linked architecture and deployment evidence. Model facts are publisher-reported or attributed to runtime maintainers; no independent GPU benchmark was run.
AI Infra Interviews, “LFM2.5-24B-A2B: release, architecture and deployment”, checked 2026-11-16. Preserve this date and the exact checkpoint when citing.