Self-hosted weights
Free
Apache-2.0; compute costs separate
- Included scope
- Clef and Clef-flash decision models
Structured decision models for agent workflows · Official-source review 2026-10-09
Clef and Clef-flash classify text or image context into schema-defined choices, scores and probabilities. Cloudflare offers hosted Workers AI models and Apache-2.0 weights, with an FDE-assisted fine-tuning service described separately.
Relevant to developers validating classifications and probability calibration against their own labeled cases before automated routing.
Hosted Clef/flash rates and FDE service pricing unconfirmed; unrelated Cloudflare Plans tag page is not model pricing. Model advertises 64K context and vision input; schema-bound output is not general text generation. Clef favors precision, flash latency per vendor; published benchmark/latency comparisons are vendor experiments, not our tests. FDE service available by discussion; self-service RL platform described as future/work in progress. Own-hosted weights do not make compute free. Request privacy claims need contract verification. No weights downloaded, model run, classification validated or fine-tuning performed.
Define valid schema choices, test labeled examples and failure paths, and require review or safe fallback before high-consequence routing.
Hosted Clef/flash rates and FDE service pricing unconfirmed; unrelated Cloudflare Plans tag page is not model pricing. Model advertises 64K context and vision input; schema-bound output is not general text generation. Clef favors precision, flash latency per vendor; published benchmark/latency comparisons are vendor experiments, not our tests. FDE service available by discussion; self-service RL platform described as future/work in progress. Own-hosted weights do not make compute free. Request privacy claims need contract verification. No weights downloaded, model run, classification validated or fine-tuning performed.
Official product evidence reviewed on 2026-10-09; product artwork inspected visually. No hands-on vendor application test was performed. Application performance, security claims and plan enforcement were not independently tested.
What it can do
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| Product workflow | Clef and Clef-flash classify text or image context into schema-defined choices, scores and probabilities. Cloudflare offers hosted Workers AI models and Apache-2.0 weights, with an FDE-assisted fine-tuning service described separately. | Facts sourced | 2026-10-09 |
| Pricing and restrictions | Hosted Clef/flash rates and FDE service pricing unconfirmed; unrelated Cloudflare Plans tag page is not model pricing. Model advertises 64K context and vision input; schema-bound output is not general text generation. Clef favors precision, flash latency per vendor; published benchmark/latency comparisons are vendor experiments, not our tests. FDE service available by discussion; self-service RL platform described as future/work in progress. Own-hosted weights do not make compute free. Request privacy claims need contract verification. No weights downloaded, model run, classification validated or fine-tuning performed. | Facts sourced | 2026-10-09 |
Understand the total cost
Self-hosted weights
Free
Apache-2.0; compute costs separate
Hosted Workers AI
Not listed
Model rates and allowances unconfirmed
FDE fine-tuning
Not listed
Service quote; self-service platform future
chat-recall stores coding conversations for full-text and knowledge-graph retrieval. It offers MCP access, analytics and code-related findings, with hosted and self-hosted deployment options.
Explore toolRunInfra provides hosted open models through OpenAI- and Anthropic-compatible APIs. It offers prepaid token billing and workspace coding plans, with session-aware caching and separately reported cached-token usage.
Explore toolAdapt provides a configurable runtime for company agents with shared knowledge, connected tools, model routing and isolated task sandboxes. It offers a self-hosted open-source runtime, managed service and enterprise deployment options.
Explore toolThe practical questions
No. This listing is based on dated official-source evidence and a visual artwork check. Unconfirmed costs and restrictions are identified explicitly.
Clef and Clef-flash classify text or image context into schema-defined choices, scores and probabilities. Cloudflare offers hosted Workers AI models and Apache-2.0 weights, with an FDE-assisted fine-tuning service described separately.
Apache-2.0 model weights available to run yourself; hardware, inference and hosting costs separate. No model-specific free hosted allowance confirmed.. This record lists ongoing free access; check the plan limits before starting.
No paid monthly price is listed; this record treats the product as free to start. See the plan cards for entitlements, billing commitments and seat minimums.
Workers AI inference endpoint and Jev-compatible typed decision schema documented. Hosted API account/authorization required.. API access and subscription access may have different terms; consult the linked sources.
Hosted Clef/flash rates and FDE service pricing unconfirmed; unrelated Cloudflare Plans tag page is not model pricing. Model advertises 64K context and vision input; schema-bound output is not general text generation. Clef favors precision, flash latency per vendor; published benchmark/latency comparisons are vendor experiments, not our tests. FDE service available by discussion; self-service RL platform described as future/work in progress. Own-hosted weights do not make compute free. Request privacy claims need contract verification. No weights downloaded, model run, classification validated or fine-tuning performed.
No retained pricing changes yet. A current price alone does not establish a historical trend.
Current official product source read individually; product artwork inspected visually.
Read original source ↗Official product page reviewed for disclosed costs; unverified prices and allowances are identified in the listing.
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.