See how ChatGPT, Perplexity and Google AI Overviews describe you today
Free AI Visibility AuditVenice Uncensored is an instruction-tuned text model from Venice.ai, built with dphn.ai as a fine-tune of Mistral-Small-24B-Instruct-2501 and released as the Dolphin Mistral 24B Venice Edition. It is positioned as an uncensored assistant, meaning it is tuned to follow instructions without the refusal layer most hosted models carry.
01 / decision
Each figure sits next to the middle of the field, so it can be read as dear or cheap, wide or narrow, rather than floating on its own. Compared against the 126 models we track, not against an absolute standard.
Capability
Not measured
No published benchmark scores yet
Input / 1M tokens
$0.20
field median $0.30
Output / 1M tokens
$0.90
field median $1.25
Context
128K
field median 524K
02 / overview
Four questions, answered separately, because somebody arrives at one of them rather than at the top of the page.
The job it was built for, and the job it is not for.
Venice Uncensored is a text-only instruct model derived from a mid-sized Mistral base, fine-tuned so that it answers rather than declines. The job it was built for is straightforward instruction following on prompts that a mainstream assistant would refuse or soften. It is not a reasoning flagship, not multimodal, and not the model to reach for when you want a provider's safety behaviour doing work on your behalf.
Where it sits in its line, and when a sibling is the better pick.
Venice Uncensored first appeared on our tracker on 9 July 2025. Within Venice's line-up it is the permissive option rather than the capable one, so it earns its place when a task genuinely stalls on refusals, for example security research, fiction with adult themes, or content moderation work that requires handling the material itself. For general product copy, customer-facing chat or anything where a compliance team will read the logs, a mainstream frontier model is the better pick.
The API, the apps it powers, and what its limits let you do.
Access is through Venice's API, and the model also sits behind the Venice consumer chat app, which is the main place people meet it. Its context window comfortably holds long documents, transcripts or a lengthy chat history, while the output cap keeps single responses to something closer to a long article than a full report, so longer deliverables need chunking. Text in, text out, with no image or audio handling.
What changes because this exists, or why it does not.
What changes is availability rather than capability. Teams that previously ran a permissive open-weights model on their own hardware to avoid refusal behaviour can now get the same posture from a hosted endpoint at a small-model price, which removes the GPU hosting problem from that workflow. On raw quality it behaves like the 24B Mistral it is fine-tuned from, so nothing here moves the frontier.
03 / evidence
Split by category, so a strong number never hides a thin evidence base. Verified means we read it on the benchmark's own published results; a provider's claim about its own model is shown and labelled rather than dropped.
No published benchmark scores for this model yet.
We publish a score only where we can link the result to where it was published. Until a benchmark result for this model exists in a source we read, this section stays empty rather than being filled with a provider's marketing figure.
How we decide what counts as evidence04 / ledger
Every published row, grouped by category, each compared with the best published score on the same benchmark. 'Is 64% good' is a question nobody can answer; '26 points behind the leader' is one anybody can.
Nothing in the ledger yet.
Each row here carries a score, the benchmark it came from, what the leading model scored on the same test, and a link to the published result. Rows appear as results are published and read.
How we decide what counts as evidence05 / capability
Where this model is strong, and against how many peers. Ranks are against models with evidence in that category, not against everything we track: ranking against models nobody tested would rank who published, not who is better.
No category scores to shape yet.
A category score is the weighted mean of the benchmarks published for it. With no published rows there is nothing to average, and an empty chart drawn at zero would say something false.
How we decide what counts as evidence06 / cost
List API rates as last read from the provider, with the source on every row, plus every change we have recorded since we started tracking it.
| Charge | Price | Unit | Read on | Source |
|---|---|---|---|---|
| Input | $0.20 | per 1M tokens | 2026-09-24 | Check |
| Output | $0.90 | per 1M tokens | 2026-09-24 | Check |
It sits at the cheap end of the market, priced like other small open-weights models rather than like a frontier system, with output costing a few times more than input as usual. For high-volume text work the cost is close to negligible, and the practical constraint is the output length per call rather than the bill.
| Workload | Input / month | Output / month | Cost |
|---|---|---|---|
| A small product team | 20M tokens | 5M tokens | $8.50 |
| A busy support assistant | 200M tokens | 40M tokens | $76.00 |
| A document pipeline | 1000M tokens | 100M tokens | $290.00 |
List API rates, no caching and no batch discount, which both providers offer and which change the answer a great deal. Treat these as the ceiling, not the bill.
07 / specs
As listed by the provider's own catalogue and re-read every few hours. Anything absent is absent there too.
| Context window | 128,000 tokens |
|---|---|
| Maximum output | 8,192 tokens |
| Modalities | text |
| Released | 09/07/2025 |
| Status | Current |
| Catalogue identifier | cognitivecomputations/dolphin-mistral-24b-venice-edition |
08 / lineage
What this model replaced, what replaced it, and what else its provider has in the field.
No other Venice models published here yet.
Model pages are created as models appear in a provider's catalogue, so this fills in as the provider ships.
How we decide what counts as evidence09 / line
Every model in this family in release order, so a page from eight months ago says in one glance that two newer ones exist.
Nothing else in this line yet.
A line is worked out from the naming and the release dates across every model page we hold. It fills in as the provider ships successors, or as we pick up the models that came before this one.
How we decide what counts as evidence10 / notes
What this model changes for a brand trying to be cited in AI answers, and every change we have logged since it launched.
What it changes for you
For a SaaS brand tracking AI visibility, this model changes very little, because it is not wired into the AI search surfaces that generate buyer answers. It matters mainly as a signal that the permissive tier of the market is now hosted and cheap, so if your category attracts adversarial prompting, someone can now probe your product and your competitors without a refusal getting in the way. Monitor it the way you would monitor any open-weights model, occasionally and for reputation reasons, not as a citation channel.
Most buyers encounter Venice Uncensored inside Venice's own chat app, where it is the default permissive assistant, or via the Venice API if they are building on it. It does not sit behind any mainstream AI search surface, so it is not a model that shapes what buyers see when they research vendors.
Nothing published here yet. Changes appear within hours of a provider announcing them.
11 / questions
The things people ask about this model, answered from what is on this page rather than from anywhere else.
Models change what gets cited. We measure whether AI answers name your brand or your competitors across every assistant, and show you what to change.