AllocateLFM2.5 8B A1B vs Inkling FP4
LFM2.5 8B A1B is not currently in the Allocate serving catalog, so this page lists no prices for it: every price on this site comes from the live catalog.
L LFM2.5 8B A1B
Inkling FP4
Inkling FP4LabLiquid AIThinking Machines
AccessNot served on AllocateOpen weights
Context windown/a512K tokens
List price, inputNot served$1 / M tokens
List price, outputNot served$4.05 / M tokens
Cached inputn/a$0.17 / M tokens
LicenseNot listedApache 2.0
Fine-tunableYesYes
Specifications and provider list prices from the Allocate catalog, checked 2026-09-12.
Choose LFM2.5 8B A1B for
- High-volume extraction
- Latency-sensitive routes
- Edge deployments
Choose Inkling FP4 for
- Fine-tuning under a permissive license (Apache 2.0)
- Published cached-input pricing ($0.17 per M tokens)
Common questions
Which has the bigger context window?
Inkling FP4: 524,288 tokens (512K) against an unlisted window for LFM2.5 8B A1B.
Can I fine-tune LFM2.5 8B A1B or Inkling FP4?
Both publish open weights (LFM2.5 8B A1B: Not listed; Inkling FP4: Apache 2.0), so both can be fine-tuned. On Allocate the trained weights stay inside your boundary and belong to you.
Related comparisons
LFM2.5 8B A1B vs GLM 4.7 FP8Inkling FP4 vs GLM 4.7 FP8LFM2.5 8B A1B vs Deepseek V3.1 NVFP4Inkling FP4 vs Deepseek V3.1 NVFP4LFM2.5 8B A1B vs Meta Llama 3.3 70B Instruct TurboInkling FP4 vs Meta Llama 3.3 70B Instruct Turbo
Run the numbers on your workload
Or do not choose. On Allocate a route name is the contract: point yours at one model today, swap to the other tomorrow, and compare them on your live traffic with per-token metering.