SubQ Snap
Quadrant scores
See the full quadrantScored with the public quadrant rubric (maturity × capability; bubble = adoption). Revise as evidence lands.
- Maturity5.3/10
Hosted Decisions API, closed weights, no SLA. On 11 Nov 2026 the id became subq-snap-1.2; the docs say 1.1 and the preview id now route to it. Docs price: $0.04/1M input, output free.
- Capability7.5/10
JevBench (17 Oct) Intelligence 72.3 on the preview row; p50 0.47 s. Noul, choice and score over 384K text and images. AA v2.1 (12 Nov): Snap 1.2 is 72.58, second among decision models, above Jev 1.14 at 66.23. Accuracy 9 and capability 7.5 unchanged.
- Adoption4/100
Launch post ~3k likes and 4.4M views (13 Oct); 100M free tokens for the first 10,000 sign-ups. No named production users yet.
Vendor claims
- Snap 1.1: +7.2 over Jev 1.14 on 16 enterprise benchmarks; When2Call 75.2 → 96.2; ECE 0.108 → ~0.014; confident OOD hallucinations 99.3% → 12.5%[Vendor claim — not independently verified]
- Decision Index 0.2.1 61.7 vs Jev 57.9; leads Jev on 13 of 16 enterprise DI benchmarks (67.6 vs 61.6 average); Decision LCR 74.0 vs Jev chunked 48.0; Decision RULER 96.3–99.8 up to 1M tokens[Vendor claim — not independently verified]
- Latency 80 ms minimal context, 250 ms at 32K, 660 ms at 100K, 2.5 s at 384K[Vendor claim — not independently verified]
- Snap 1.2 scores 73, tied for first, on Artificial Analysis Decision Index v2.1; option-order invariance is nearly perfect; the open part of the HF Decision Index scores 69.05.[Vendor claim — not independently verified]
News covering this model
Board audit of 17 Oct 2026: see the “Board audit” section below; where it conflicts with older board statements on this page, it wins.
SubQ Snap 1.2 (11 Nov 2026)
Where this section conflicts with the 1.1 model id below, this section wins. The capability score does not move.
Subquadratic released SubQ Snap 1.2 on 11 Nov 2026 (post). The models page, read 12 Nov, lists subq-snap-1.2 as the default decision id and subq-snap-latest as its alias. subq-snap-1.1 and subq-snap-preview were retired that day and route to 1.2. Price on that page is still $0.04 per 1M input tokens, output free. Weights stay closed. Question types stay noul, choice and score. Text and images.
Artificial Analysis Decision Index v2.1, read 12 Nov from the live page. Among decisionModel rows, Snap 1.2 is index 72.58 (162 ms per task), second behind Drex Max v1.1 at 72.62. Jev 1.14 is 66.23. The same figures were on the page on 11 Nov. The post’s “73, tied for first” does not match this reading (UnverifiedClaim). Language-model rows on that page are not part of this order.
The post’s LongDecisionBench numbers and the claimed 69.05 on the open part of the Hugging Face Decision Index are Subquadratic’s. Snap is not a row in the Decision Index file generated 11 Nov (297 entries).
Score unchanged: 7.5. Accuracy stays 9 under the 24 Oct decision (Artificial Analysis above Jev, plus the JevBench preview row at Intelligence 72.3). That preview row is not a 1.2 run. See news.
SubQ Snap 1.1 (22 Oct 2026)
Prices and board status below this section are superseded where they conflict.
Subquadratic released SubQ Snap 1.1 on 22 October 2026 (post; @alex_whedon, 16:48 BRT). Model id subq-snap-1.1 on POST https://api.subq.ai/v1/systemone; still closed weights. Price drops 20% to $0.04 per 1M input tokens, output free; context stays at 384K tokens.
Independent board. Artificial Analysis puts Snap 1.1 at 65 on its Decision Index v1.0, tied for first with d1-ultra and Elephant 9B (mammoth-1 family), 8 points ahead of Jev 1.14 (57). AA attributes the result to a low hallucination rate: non-hallucination 86% (Elephant 9B 93%, d1-ultra 85%), accuracy 20% on its knowledge questions, 18% on HLE-MultiChoice, $30 per 1M tasks. Snap 1.1 is not on the Decision Index or JevBench yet (JevBench still shows the preview).
Subquadratic’s numbers (UnverifiedClaim): +7.2 points over Jev 1.14 across 16 enterprise benchmarks (+5.5 over Snap 1.0); When2Call 75.2 → 96.2; expected calibration error 0.108 → ~0.014 on held-out problem families; confident hallucinations on out-of-distribution queries 99.3% → 12.5%.
Update 24 Oct: the Snap 1.1 post also lists HF Decision Index 0.2.1: 65.7 (Jev 1.14 60.6, Snap 1.0 61.8). That is Subquadratic’s own run on an old edition, not a board row (UnverifiedClaim). The only independent evidence for 1.1 is still the Artificial Analysis 65 above. Score applied: capability 7.5.
Score (approved 24 Oct): capability 7.3 → 7.5 = 9 × 0.30 + 5 × 0.25 + 7 × 0.25 + 9 × 0.20 = 7.50. Accuracy 9: two sources above Jev, Artificial Analysis 65 (1.1) and the JevBench preview row 72.3 (63.6 for Jev), counted together by Fabricio’s decision. Maturity 5.3 and adoption 4 unchanged. Working in docs/06-quadrante.md.
Read this first. Every number below is Subquadratic’s (UnverifiedClaim). The 61.7 is on Decision Index 0.2.1, an older edition than the current 0.3 board, so it can’t be ranked against 0.3 scores. Snap is not on the 0.3 board.
SubQ Snap (launch post, 12 Oct 2026) is a hosted decision model built on Subquadratic Sparse Attention (SSA), which the company says keeps compute per token constant as context grows. A decision head returns probabilities over the allowed answers, so nothing is generated. It accepts text and images in the same context, up to 384K tokens, about 12× Jev’s window by the vendor’s count. Subquadratic pitches it for routing, tool picking, checking and stopping inside agent loops where the evidence is long: a codebase, a contract and its exhibits, a long tool history.
Price: $0.05 per million input tokens; output is free. Availability: public preview, first come first served; new users may be waitlisted when capacity runs out.
Vendor results (UnverifiedClaim)
| Benchmark | Snap | Jev |
|---|---|---|
| Decision Index 0.2.1 | 61.7 | 57.9 |
| 16 enterprise DI benchmarks (average) | 67.6 | 61.6 |
| Decision LCR (72K–115K, one-shot vs chunked) | 74.0 | 48.0 |
| Bespoke Nimble Suite 13 | 82.9% | 76.0% |
Snap loses to Jev on When2Call (−7.6), API-Bank (−1.6) and ACOS (−0.6). Decision RULER and Decision LCR are Subquadratic’s own conversions of RULER and AA-LCR to multiple choice, not public boards.
Score working (13 Oct 2026)
- Maturity = 6 × 0.30 + 7 × 0.25 + 3 × 0.25 + 5 × 0.20 = 1.80 + 1.75 + 0.75 + 1.00 = 5.30 → 5.3
- Capability = 9 × 0.30 + 5 × 0.25 + 7 × 0.25 + 7 × 0.20 = 2.70 + 1.25 + 1.75 + 1.40 = 7.10 → 7.1
- Adoption = 9 × 0.35 + 2 × 0.35 + 1 × 0.30 = 3.15 + 0.70 + 0.30 = 4.15 → 4
Calibration 5: no ECE or reliability data. Latency 7: detailed vendor figures by context length. Accuracy 7: a vendor DI run with a published method, counted as evidence per Fabricio’s rule but still unverified and on the old 0.2.1 edition.
Board audit (17 Oct 2026)
Checked against the official boards on 17 Oct 2026: Decision Index 0.3.2 (data of 16 Oct, 19:00 UTC), JevBench and ImageJevBench live. The boards are separate sources and are never averaged. Older board statements and rankings further down this page are superseded where they conflict with this section.
| Board | Row | Result |
|---|---|---|
| JevBench API (live 17 Oct) | SubQ Snap | Capability 81.5; Intelligence 72.3; calibration 90.7; p50 0.47 s; $0.065/1k |
Board-status statements on this page to recheck against the table (may be superseded):
- “The 61.7 is on Decision Index 0.2.1, an older edition than the current 0.3 board, so it can’t be ranked against 0.3 scores.”
Score update (audit, 17 Oct 2026). Capability 7.1 → 7.3 = types × 0.30 + calibration × 0.25 + latency × 0.25 + accuracy × 0.20 = 9 × 0.30 + 5 × 0.25 + 7 × 0.25 + 8 × 0.20 = 7.30 → 7.3. Types, calibration and latency come from the earlier working (9 / 5 / 7); only accuracy is re-derived from the board (8; was 7): JevBench Intelligence 72.3 ÷ 63.6 × 8 = 9.1, half-up, capped at 8. Maturity and adoption unchanged.
