smsh segments enterprise prompts by semantic region, freezes behavioral invariants across turn boundaries, attests fidelity at the answer level, and amortizes post-quantum signatures across an inference session. We bill net of savings. The customer never pays more than the compression saves.
Free demo. No auth, no card. Paste a prompt and read the receipt.
No auth, no card. Paste up to 4,000 characters. We return the compressed payload, the ratio, dictionary hits, and the latency. The exact same endpoint your application calls.
Fidelity contract. FIPS-conformant PQ. Honest ratio bands measured tonight. Kill conditions we ship instead of press releases.
Not a hash. Not a vendor claim. A bound chain from compressed prompt to verified answer to post-quantum signature. Five layers, each one verifiable independently.
The pruner never splits a BPE token. Mask is constructed via HuggingFace Tokenizers' offset_mapping; sub-token positions are forced non-prunable.
Instruction, schema, retrieved context, user input, examples, output spec each get their own compression policy. Schema fields and output specs are pass-through by default.
Negations, numbers, named entities, schema fields, tool names, refusal triggers are non-prunable in turn one and stay non-prunable through turn N. Multi-turn cap default 2.5x.
7B fine-tuned NLI, async on a sample of compressed inferences, calibrated to under 1% false-pass on held-out probes. Flag triggers automatic customer billing credit.
Ed25519 inner per inference (64 bytes). ML-DSA-65 outer over Merkle root of inference receipts per session (3309 bytes amortized). ML-KEM-768 wraps the session key (1088 bytes ct).
Hit the self-test endpoint and read the byte sizes and latencies for ML-DSA-65 and ML-KEM-768 against the FIPS-specified values.
# Live FIPS-conformance self-test, no auth, returns byte sizes + latencies curl https://receipts.thehiveryiq.com/v1/smsh/pq/selftest # Fetch a fresh sample ML-DSA-65 keypair for demo verification curl https://receipts.thehiveryiq.com/v1/smsh/pq/keys/sample
Single-number compression claims are how vendors get caught. We publish CDF by prompt length × task × turn count, and we'll send you the bench repo. Below are the bands we stand behind.
| Workload | Band | Bar |
|---|---|---|
| Broad single-prompt, task-preserving | 2x to 5x | LLMLingua-2 sidecar target at ≤1.5% accuracy loss (v1.5) |
| Repetitive long-context tenant traffic (batched RAG, CASB templates) | 4x to 8x | Tenant dictionary + region segmentation engaged |
| Multi-turn (3+ turns) with invariant freeze | 1.5x to 2.5x | Hard cap at 2.5x; below 1.5x if fidelity drop > 10% |
| Short prompts (< 150 tokens) | pass-through | Router skips compression. No overhead. |
| "8x broadly" | never | We don't claim this. Anyone who does is selling the lie. |
| "100% HumanEval" | never | We publish range, not headline. |
| "Sub-millisecond compression" | never | Production latency is 1 to 15 ms. The lie is the qualifier. |
| Workload | Ratio | Latency | Invariants |
|---|---|---|---|
| RAG repetitive (paraphrased context, 237 t) | 1.58x | 4 ms | all preserved |
| RAG paraphrased (3 facets of one topic, 156 t) | 1.23x | 3 ms | all preserved |
| Verbose-fillers (customer ticket rewrite, 159 t) | 1.14x | 2 ms | all preserved |
| CASB-mixed (DLP policy + entity-heavy, 207 t) | 1.04x | 2 ms | PII, refusal, deontics preserved |
| Schema-heavy (JSON output spec, 255 t) | pass-through | 2 ms | schema frozen |
| Chat short (< 80 t) | pass-through | 0.1 ms | no overhead |
v1 is the deterministic scaffold. v2 is the engine. A 4-layer transformer (≈20M params, int8 ONNX, ≈30 MB on disk), trained on a corpus of CASB, RAG, agent-context, and conversation prompts labeled by LLMLingua-2, then conditioned on the region type emitted by v1's segmenter. The teacher labels everything. The student runs on CPU.
| v2 target | Commitment | Why we can promise it |
|---|---|---|
| Mean compression across CASB / RAG / agent | 3.0 to 3.5x | LLMLingua-2 teacher hits 2.5 to 3.5x. Region-conditioning lifts it further on workloads it was never trained for. |
| Warm CPU latency, 1K tokens | < 50 ms | The 4-layer student is 30x smaller than the BERT-base teacher. ONNX int8 halves it again. |
| Invariant preservation (PII, numbers, deontics) | ≥ 99.5% | The hard veto from v1's invariant set is added to the student's keep mask. The teacher can't override it. |
| Status | training now | 50K-prompt typed corpus generated. Teacher labeling on GPU spot starts tonight. Student trains over the next 24 to 72 hours. When it clears the bars above, it ships. When it doesn't, it doesn't. |
If any of these fires during the v1 build, we don't claim what we can't deliver. We label honestly or we don't ship the feature.
Endpoints to call. The offer (we only bill what we save). Questions answered.
# Free Tier0 compression demo, returns ratio + receipt curl -X POST https://receipts.thehiveryiq.com/v1/smsh/compress/demo \ -H "Content-Type: application/json" \ -d '{"text": "<your prompt>", "mode": "smsh-v1"}' # Verify FIPS conformance live curl https://receipts.thehiveryiq.com/v1/smsh/pq/selftest
Under NDA. We mine your tenant dictionary. We measure your actual ratio on your actual traffic. If we beat your status quo cost-per-inference at no worse than 1.5% accuracy delta, you pay our net-savings rate. If we don't, you don't.
No deck. No POC fee. The deployment is live, the endpoints respond, the PQ self-test passes FIPS 203/204 conformance today. Read the receipt yourself.
Reach out via X or hit the demo endpoint above. The bench repo is open-source.