Head to head · Decision models · October 2026 research run
GLiClass vs RightNow System Zero
RightNow System Zero scores 66.8 (B) on agent readiness against GLiClass's 50 (D), and leads in 6 of 7 scored categories. GLiClass leads on payments & pricing. Both do decision models.
Best decision models for AI agents · All 220 decisions comparisons
Which one, for what
GLiClass D
Best for Topic, intent and sentiment routing over a known label set on the owner's own CPU or GPU, where many labels must be scored at once.
Ahead on
- Payments & pricing, 60 against 40
Also in its favour
- No key needed to call it
- Open source
Watch for
No calibration evidence. The cards report F1 only, and the docs tell users to calibrate thresholds on their own traffic
Best for Typed decisions with per-option probabilities over text, JSON, images, audio and PDFs, including on streaming state
Ahead on
- Reliability, 50 against 43
- Schema & documentation, 83 against 49
- Agent ergonomics, 86 against 60
- Security & auth, 70 against 38
- Maintenance & community, 68 against 44
- Transparency & trust, 70 against 65
Also in its favour
- A hosted endpoint, with nothing to install
- Free to start without a card
Watch for
Standard context is 38,000 tokens. The 64k, 128k and 1M bands are held, and video is a preview
Score by category
| Category | Weight this run | GLiClass | RightNow System Zero | Edge |
|---|---|---|---|---|
| Reliability | 16%20 | 43 | 50 | RightNow System Zero +7 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13%16.2 | 49 | 83 | RightNow System Zero +34 |
| Agent ergonomics | 13%16.2 | 60 | 86 | RightNow System Zero +26 |
| Security & auth | 14%17.5 | 38 | 70 | RightNow System Zero +32 |
| Payments & pricing | 10%12.5 | 60 | 40 | GLiClass +20 |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7%8.8 | 44 | 68 | RightNow System Zero +24 |
| Transparency & trust | 7%8.8 | 65 | 70 | RightNow System Zero +5 |
| Negative events | ≤15 | 0 | 0 | |
| Total | D 50/100 | B 66.8/100 |
Facts side by side
| Fact | GLiClass | RightNow System Zero |
|---|---|---|
| Kind | Model API | Model API |
| Vendor | Knowledgator | RightNow AI, Inc. |
| Hosted endpoint | no (local only) | https://api.rightnowai.co/v1/system-zero/decide |
| Transports | HTTP | HTTP, websocket |
| Auth | None | API key |
| Pricing | Free | Pay per use |
| x402 | no | no |
| Licence | Apache-2.0 (library and the model weights we checked) | Proprietary hosted model under RightNow AI terms of use; the SDKs are Apache-2.0 |
| Read-only variant documented | no | no |
| llms.txt | no | yes |
| Last release | 2026-07-21 | 2026-10-27 |
| Terms last updated | no document linked | no date given |
| Privacy policy last updated | no document linked | no date given |
| Customer content may train models | yes | |
| Terms restrict automated access | yes | |
| Terms restrict benchmarking | not found in the text | |
| Terms or service can change without notice | not found in the text | |
| Arbitration or class-action waiver | not found in the text | |
| Popularity | 555 stars, 13k PyPI/wk | none |
Verdicts
GLiClass
An Apache-2.0 classifier that scores a whole label set in one encoder pass on the owner's hardware, with single-label, multi-label, hierarchical and few-shot modes. It returns label scores with no calibration claim, the bundled server has no authentication, and the last three test runs on the main branch, on 24 September 2026, failed.
RightNow System Zero
A well-documented decision API with an OpenAPI file, idempotent retries, a detailed error catalogue, streaming over WebSocket and published per-unit prices with USD 5 of starting credit. Standard context is 38,000 tokens; the 64k, 128k and 1M bands are held and video is a preview. The latency and benchmark figures are RightNow's own, and several of its pages disagree.
Before you call either
GLiClass
- Pass
--host 127.0.0.1topython -m gliclass.serve, or put the port behind your own gateway. The server checks no credential - On a machine without a GPU add
--device cpu --dtype float32 --num-gpus-per-replica 0. The default configuration expects CUDA - Send one text a request to
POST /gliclass. An array intextsis cut to its first item without an error - Set
multi_labelto false for one label from a set. The default scores each label independently, so scores do not sum to 1 - Keep text plus labels under the pipeline's 1,024-token
max_length, or useZeroShotClassificationWithChunkingPipeline. Longer input is truncated silently
RightNow System Zero
- Send audio only as inline 16,000 Hz WAV. The FLAC, MP3, OGG and MP4 audio named on the blog are refused with media_unsupported.
- Keep state under 38,000 tokens. Long-context requests return 404 long_context_not_available, which never succeeds on retry.
- Reuse the same Idempotency-Key and body on retries, and stop when error.retryable or x-should-retry is false.
- Check every entry in answers even on HTTP 200, since one question can be refused while the others are answered.
- Treat probabilities as uncalibrated while calibrated is false, and pin system-zero-2.0.0 when answers must not move with the alias.
Questions
Which is better for AI agents, GLiClass or RightNow System Zero?
RightNow System Zero scores 66.8 (B) on agent readiness against GLiClass's 50 (D), and leads in 6 of 7 scored categories. GLiClass leads on payments & pricing.
Do GLiClass and RightNow System Zero need an API key?
GLiClass needs no key. RightNow System Zero needs an API key.
Can an agent call GLiClass and RightNow System Zero without installing anything?
No hosted endpoint is listed for GLiClass. RightNow System Zero has a hosted endpoint at https://api.rightnowai.co/v1/system-zero/decide.
Are GLiClass and RightNow System Zero open source?
GLiClass is open source (Apache-2.0 (library and the model weights we checked)). No open-source release is listed for RightNow System Zero.
Other comparisons with GLiClass or RightNow System Zero
- Celeris-1 Decision vs GLiClass
- Celeris-1 Decision vs RightNow System Zero
- Clef vs GLiClass
- Clef vs RightNow System Zero
- Laya vs GLiClass
- Laya vs RightNow System Zero
- Decider vs GLiClass
- Decider vs RightNow System Zero
- Decision 3.0 vs GLiClass
- Decision 3.0 vs RightNow System Zero
- GLiClass vs Granite Switch 4.1
- GLiClass vs Kev
- GLiClass vs Liquid d1
- GLiClass vs Microsoft-Decision-1
- GLiClass vs Drex 1.5
- GLiClass vs OpenAI Decisions API
- GLiClass vs pplx-decider
- GLiClass vs pplx-decider-flash
- GLiClass vs ScaleDown
- GLiClass vs SPEV
- GLiClass vs kettei
- GLiClass vs Strands Decider 2B
- GLiClass vs Jev
- GLiClass vs Vela 2.0
- Granite Switch 4.1 vs RightNow System Zero
- Kev vs RightNow System Zero
- Liquid d1 vs RightNow System Zero
- Microsoft-Decision-1 vs RightNow System Zero
- Drex 1.5 vs RightNow System Zero
- OpenAI Decisions API vs RightNow System Zero
- pplx-decider-flash vs RightNow System Zero
- pplx-decider vs RightNow System Zero
- RightNow System Zero vs ScaleDown
- RightNow System Zero vs SPEV
- RightNow System Zero vs kettei
- RightNow System Zero vs Strands Decider 2B
- RightNow System Zero vs Jev
- RightNow System Zero vs Vela 2.0
Machine-readable
- This page as Markdown
/compare/gliclass-vs-rightnow-system-zero.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/gliclass.json·/api/v1/tools/rightnow-system-zero.json - From a terminal
anchor compare gliclass rightnow-system-zero(the CLI) - Over MCP
compare_tools {"a": "gliclass", "b": "rightnow-system-zero"}at/mcp, no key