AllocatePearl Gemma 4 31B Instruct vs Inkling FP4
Pearl Gemma 4 31B Instruct 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.
P Pearl Gemma 4 31B Instruct
Inkling FP4
Inkling FP4LabPearl 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 Pearl Gemma 4 31B Instruct for
- Long-document reasoning
- Open-weight fine-tuning
- Mid-size general work
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 Pearl Gemma 4 31B Instruct.
Can I fine-tune Pearl Gemma 4 31B Instruct or Inkling FP4?
Both publish open weights (Pearl Gemma 4 31B Instruct: 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
Pearl Gemma 4 31B Instruct vs GLM 4.7 FP8Inkling FP4 vs GLM 4.7 FP8Pearl Gemma 4 31B Instruct vs Deepseek V3.1 NVFP4Inkling FP4 vs Deepseek V3.1 NVFP4Pearl Gemma 4 31B Instruct 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.