- An OpenRouter Responses rung (openrouter.ai origin) no longer receives include: ["websearchcall.action.sources"], which OpenRouter refuses with 400 "Invalid Responses API request" while relaying…
- A replayed reasoning item loses its output-only status, as input messages already did. OpenAI refused it with 400 "Unknown parameter: 'input[N].status'" and serves the same item without it.
- Includes 0.7.177 (#1261): a gateway reasoning carrier stays with its own tool turn when a client records a developer or system message inside it.
Experiential by Experiential Labs
Experiential is the open source, zero markup gateway for BYOK, self-hosted and 1000+ marketplace models. It learns from your traffic to cut costs, recommend better models, and train a specialized model you own.
Agent tooling for cost, context and memory
New · Added to AgentGid on Oct 21, 2026Open source · Apache-2.0Free to start
Recently added on Oct 21, 2026. The description, category and pricing were filled in from public metadata and have not been checked by an editor yet.
Price
Free (OSS)
free and open source
Popularity
#44 in Agent tooling
Gid Score 32 / 100 · #44 of 65 ranked agent tools
Trend
Gaining stars
about +11 GitHub stars a day
Development
Actively updated
latest 0.7.178, 11d ago
Community
9,972 GitHub stars
public signals
Usage data: latest readings Oct 20 to Oct 21, 2026, last checked Oct 21, 08:10 UTC. Date of each number
What Experiential does
It is free and open source (Apache-2.0); you pay only for the models it calls.
Key facts
| Maker | Experiential Labs |
|---|---|
| First released | June 2026 |
| License | Apache-2.0 |
| Main language | Python |
| Open issues and PRs | 144 |
Pricing
Open source under the Apache-2.0 licence; check the website for any paid hosted plans.
, checked Oct 21, 2026. Prices change often; confirm before you buy.
Quick answers
- Free to start?Yes
- Open source?Yes (Apache-2.0)
- Closest alternativeOpenShell · all alternatives
Experiential alternatives
Best Experiential alternatives compared · All agent tooling for cost, context and memory
| # | Agent | Gid Score | Downloads 7d | 7d | 30d | Stars | Latest release | Price | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 |
OpenShellNVIDIA |
77 | 101K | down 14.1% | up 381.4% | 16.8K+103/day | 0.1.5yesterday | Free (OSS) | |||
| 2 |
|
72 | 22.9M | up 4.3% | — | 61.7K+79/day | 1.104.5today | Free (OSS) | |||
| 3 |
NVIDIA NemoClawNVIDIA |
69 | — | — | — | 22.8K+10/day | — | Free (OSS) | |||
| 4 |
|
64 | 9.5M | down 17.1% | up 17.7% | 36.1K+48/day | 4.66.0yesterday | Free + $29/mo | |||
| 5 |
|
64 | 2M | down 19.8% | up 0.3% | 194K+350/day | 2.11.04mo ago | Free + $16/mo | |||
| 6 |
|
60 | 3.2M | down 14.5% | up 21.9% | 14.5K+21/day | 2.11.2yesterday | Free + $150/mo | |||
| No agents match that filter. | |||||||||||
Ratings from people who use Experiential
No ratings yet. Used Experiential? Rate it and say what it is good and bad at.
What changed: recent Experiential releases
80 stable releases in the last 90 days · Full changelog
- Keeps a gateway reasoning carrier with its own tool turn when a client records a developer or system message between the turn's items. Codex can append a developer reminder after a function call…
Earlier releases
- Codex records Responses items as they stream and can append a developer reminder after a function call (or a carrier) and before the rest of that turn. The replayed functioncall, developer,…
- In a run holding a carrier, an instruction recorded after a call or a carrier is moved after the rebuilt tool turn, ahead of the tool outputs; the carrier still authenticates against its sealed…
- exp.simulation.scenarios.selectauthoredscenarios adds caller-supplied scenarios (instruction, optional context, tools and a judge-only reference answer) after a completed build's mined tasks;…
- ModelEvaluationOptions(includemined=False) replays only the authored scenarios; the default replays both. Plans, quotes, judging and reports follow the choice.
- The reference answer reaches only the judge, never candidates, environments or the world model. Hosted, spend- or router-bound projects refuse a scenario swap; router optimization refuses authored…
- Includes #1258: OpenRouter-style cachedtokens for Gemini cache writes read back, plus renewable automatic caches (native 0.3.143).
- Chat {"type": "custom", "custom": {...}} declarations (text, lark or regex grammar, or no format), a custom toolchoice on both surfaces, Chat allowedtools custom entries, and custom calls and…
- Native Responses rungs receive the declarations, grammar, selector and history verbatim; admission prefers such a rung when a custom tool declares a grammar.
- Other wires serve a custom tool as a function with one string input (a grammar is stated in its description), disclosed as tools.custom->translated(customasfunctioninput) and…
- Converted history keeps one assistant turn's calls in one message; the Responses normalizer marks custom starts as custom.
- A recovered (cache-affinity) session whose warm rung is last in its stage keeps the stage's other leaves as fallbacks instead of a one-rung route, so a stall on the warm rung fails over.
- When two live rungs sealed reasoning carriers in one tool loop, the newest issuer is pinned and older-issuer carriers are dropped unrevealed (disclosure…
- A request-level prefirstbytebudgetseconds (default 90 s, zero disables) bounds everything before the caller's first byte across all attempts, so failover chains end under Cloudflare's 100 s origin…
Experiential popularity and usage data
The GitHub repository has 9,972 stars, 4,700 of them added in the last 30 days. The latest stable release is 0.7.178, published Oct 10, 2026; there were 74 stable releases in the past 30 days. It was mentioned in 8 Hacker News posts and comments over the last 30 days.
Download the raw daily series: experiential.csv
Gid Score breakdown
Adoption18
Community42
Attention7
Momentum95
Each component is scored 0 to 100 from public signals; a dash means no data for it. Methodology.
How fresh these numbers are
| Number | Latest reading | Last checked |
|---|---|---|
| GitHub stars | Oct 21, 2026 | Oct 21, 08:10 UTC |
| Website rank (Tranco) | Oct 20, 2026 | Oct 21, 08:10 UTC |
| Hacker News mentions | Oct 20, 2026 | Oct 21, 08:10 UTC |
We fetch every source once a day. Registries publish downloads and page views a day or two late, so their latest reading trails the check. A reading more than 4 days old is marked out of date; the Gid Score still uses it until a newer one arrives.
Experiential FAQ
How much does Experiential cost?
Open source under the Apache-2.0 licence; check the website for any paid hosted plans.
Is Experiential open source?
Yes. Experiential is open source, released under the Apache-2.0 license.
How popular is Experiential?
The GitHub repository has 9,972 stars, 4,700 of them added in the last 30 days. The latest stable release is 0.7.178, published Oct 10, 2026; there were 74 stable releases in the past 30 days. It was mentioned in 8 Hacker News posts and comments over the last 30 days.
What is the latest version of Experiential?
The latest stable release we track is 0.7.178, published on Oct 10, 2026.
What are the alternatives to Experiential?
The closest alternatives in the same category by Gid Score are OpenShell, LiteLLM, NVIDIA NemoClaw, Langfuse.
For the makers of Experiential badges, corrections
Add a live badge to your README or site:
[](https://agentgid.com/agents/experiential/)[](https://agentgid.com/agents/experiential/)Something wrong or outdated on this page? Send a correction with a link to the official source. More for makers.
Where these numbers come from
GitHub: experientiallabs/experiential. Site rank: experientiallabs.ai on the Tranco list. See data sources for how each one is collected.
Product facts and pricing were checked against: experientiallabs.ai, github.com.