Cost / token optimisation companies, products & suppliers
Products whose job is reducing token or inference spend. Jobs: cache and compression; spend alerts; per-feature cost allocation.
What is Cost / token optimisation?
Products whose job is reducing token or inference spend. Jobs: cache and compression; spend alerts; per-feature cost allocation.
What problems does it solve?
Token bills grow without a control loop.
Typical business use cases
- Cache and compression
- Spend alerts
- Per-feature cost allocation
Important capabilities
- Caching
- Budgets
- Allocation
What buyers should evaluate
- Whether quality is measured when cost drops
- Log contents
- Budgets
Risks and governance considerations
Aggressive caching that serves the wrong tenant's completion.
Procurement checklist
- Cache tenancy
- Quality guardrail
- Budget owners
Relevant AI Trustmark assurance
AI Trustmark independent findings appear only when an assessment or certificate exists. Category membership does not imply verification.
Products and providers
Open a product profile for its catalogue details, or use the provider link for the company profile. Claimed listings appear first, then by name. Payment never changes the order.
By FriendliAI
FriendliAI is the fastest inference cloud for agents, built to run frontier open-weight models in production at scale. It delivers up to 7x faster output token speed, up to 90% low
By Helicone
Routing and monitoring for reliable AI apps - the LLMOps platform behind the fastest-growing AI companies. Helicone Cost Tracking is Helicone's platform, from Helicone, Inc. Publis
By Inferless
Blazing fast serverless GPU inference to deploy ML models. Use Inferless for scalable and effortless custom machine learning model deployment. Inferless is Inferless's platform, fr
Helping open technology projects build world class open source software, communities and companies. DeepSpeed is Linux Foundation's platform, from The Linux Foundation. Published a
By NVIDIA
NVIDIA Dynamo is an open-source, low-latency, modular inference framework for serving generative AI models in distributed environments. It scales inference across large GPU fleets
By Red Hat
Red Hat is the world’s leading provider of open source solutions, using a community-powered approach to provide reliable and high-performing cloud, virtualization, storage, Linux,
Products
Claimed products appear first. Ranking packs and payment do not change this list.
- DeepSpeed· Linux Foundation
- Friendli Dedicated Endpoints· FriendliAI
- Helicone Cost Tracking· Helicone
- Inferless· Inferless
- KServe· Linux Foundation
- llm-d· Linux Foundation
- NVIDIA Dynamo· NVIDIA
- NVIDIA TensorRT-LLM· NVIDIA
- Red Hat AI Inference Server· Red Hat
- vLLM· Linux Foundation
Related categories
Relevant procurement and assurance guides
Frequently asked questions
What is Cost / token optimisation?
Products whose job is reducing token or inference spend. Jobs: cache and compression; spend alerts; per-feature cost allocation.
What should not be listed as Cost / token optimisation?
Products whose buyer job is gateways, routers whose primary job is quality, or FinOps for generic cloud. Those belong on their own category page so search queries are not split.
Has AI Trustmark independently assessed every Cost / token optimisation supplier?
No. A category listing is descriptive. Independent assessment is shown only on company or product pages that carry Trustmark evidence.
What logging should an AI agent provide?
Buyers should be able to see who the agent acted as, which tool was called, what data was sent, what changed, and when. Logs that omit write actions or store raw customer prompts without access control are incomplete evidence.
What model or provider changes should a buyer insist on being told about?
Material change usually includes a new model family, new region, new subprocessor, new write-capable tool, or a change that affects logging, privacy or human oversight. Those changes should trigger evidence refresh rather than a silent release.
How should buyers verify where AI customer data is processed?
Ask for the named processing locations, cloud regions and any subprocessors that see prompts, files or outputs. A directory listing is not evidence of residency. Independent assessment records the locations that were in scope on the assessment date.
Does a TrustMark on one product cover the rest of the company?
No. Independent assessment is scoped to the named organisation and, where relevant, the named product. Category pages list suppliers as a topic label. They do not imply that every listed company has been assessed.
What incident-handling evidence is useful for AI suppliers?
Buyers should see how AI-specific failures are detected, contained and notified — including unsafe outputs, data leakage and unauthorised agent actions. An incident policy that never mentions models, prompts or tools is incomplete for this class of product.