Skip to content
ModelCap

Dots3-Note Preview

Official publisher releaseLimited evidenceOpen weightsFree tier6.9d #49 → #50

Dots Studio·#50 of 371 on the current board·cheaper than #49 Step 5 Preview

Dots3-Note Preview is a language model from Dots Studio. It ranks #50 of 371 on ModelCap's current board, with a thinly supported Index score of 67.8, partly measured on public benchmarks and partly modeled. Its score range of 55.3–80.3 spans positions #11–#107. It is listed free of charge from one provider. It reads up to 512K tokens of context, and its weights are openly downloadable.

Figures as of 19 Nov 2026, 05:16 UTC · download the public dataset · methodology

ModelCap Index · Current board
67.8#50
Partly measured on public benchmarks and partly modeled · range 55.3–80.3
Evidence as of 18 Nov 2026, 14:36 UTC
Market Gravity
31.6
Secondary market signal
Input
Free
per 1M tokens
Output
Free
per 1M tokens
Context
512K
tokens
Providers
1
endpoint

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
AtlasCloudfp8FreeFree512Kfp8100.0%

Reported evaluation leads (12)

Not independent measurement

Structured rows reported in this exact model repository at the pinned revision below. They are the publisher talking about itself: never measured evidence, never a score input on their own. Rows marked as used were laddered against independently measured peers inside the launch-card channel of this model's modeled placement, with the replay-measured optimism haircut applied; the others played no part.

Dataset / taskMetricValueRevisionProvenance
benchflow/skillsbench
skillsbench_v1_1
—52.8
Unpinned
not published
Reported
Not used in launch estimate
cais/hle
hle
—52.6
Unpinned
not published
Reported
Not used in launch estimate
claw-eval/Claw-Eval
general
—73.4
Unpinned
not published
Reported
Not used in launch estimate
harborframework/terminal-bench-2.1
terminalbench_2_1
—75.1
Unpinned
not published
Reported
Not used in launch estimate
hkust-nlp/Toolathlon
toolathlon_verified
—55.6
Unpinned
not published
Reported
Not used in launch estimate
internlm/WildClawBench
overall
—61.7
Unpinned
not published
Reported
Not used in launch estimate
mercor/apex-agents
apex-agents
—30.8
Unpinned
not published
Reported
Not used in launch estimate
MME-Benchmarks/Video-MME-v2
video-mme-v2
—39.3
Unpinned
not published
Reported
Not used in launch estimate
MMMU/MMMU_Pro
mmmu_pro_standard_10_options
—79.1
Unpinned
not published
Reported
Not used in launch estimate
ScaleAI/SWE-bench_Pro
SWE_Bench_Pro
—61
Unpinned
not published
Reported
Not used in launch estimate
SWE-bench/SWE-bench_Multilingual
swe_bench_multilingual_%_resolved
—75.7
Unpinned
not published
Reported
Not used in launch estimate
SWE-bench/SWE-bench_Verified
swe_bench_%_resolved
—78.4
Unpinned
not published
Reported
Not used in launch estimate

ModelCap Index · Current board

Methodology
67.8
Partly measured on public benchmarks and partly modeled
Current board
#50 of 371

Limited evidence. Measured on only 1 public benchmark, with limited support, so the score is less certain than a broadly measured one. What this label means

Score range
55.3–80.3
Rank range
#11–#107
Evidence as of
18 Nov 2026, 14:36 UTC
Benchmarks used
ARC-AGI-2 Verified
Technical details
Public rank basis
Measured · blended (measured)
Evidence label
Measured · blended · 10 reported rows against measured corpus ladders (69%); global-corpus-prior over 214 held-out anchors (18%); 1 specialist-board observation at 2.1% support (13%); shrunk 3.4 toward the measured corpus
Index support
2.1% · weak
Identity binding
Exact catalogue product · dots-studio/dots-studio/dots-3-note-preview · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata dots-studio/dots3-note-prev@1e1e7b0cd37a3a48a6c8d7fa55d5f9d14377006b (not the evaluation revision)
Index observations used
1
Qualified Index evidence point
67.8
How this position was produced · fused launch placement

No general-family board lists this model yet. Its point estimate is the precision-weighted mean of every channel that spoke for it; each channel’s share is the weight it carried. Independent measurements replace this placement as boards list the model.

Fused estimate
67.8 ± 9.8
Independent share
13%
Shrink toward corpus
-3.4
Optimism haircut
None applied
Method
precision-weighted-v1
Channels fused into this model’s launch placement
ChannelEstimateShare
Corpus ladder · not independent
10 reported rows against measured corpus ladders
72.9 ± 11.769%
Publisher corpus prior · not independent
global-corpus-prior over 214 held-out anchors
52.7 ± 22.918%
Specialist board measurement
1 specialist-board observation at 2.1% support
61.1 ± 27.613%
Bayesian Evidence Score · measured evidence input only

BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.

Observed capability
61.1
Evidence status
provisional · 2% mass
General preference
—
Coding
—
Agents & tools
—
Reasoning
61.1
Evidence breadth
3%
Evidence coverage
2%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
38.2
Liquidity (providers × uptime)
22.9
Open reach (HF downloads)
9.3
Surface (context / tools / modalities)
95.1
Freshness
69.1
  • · OpenRouter popularity #37
  • · 1 live provider
  • · HF 30d downloads 4297
  • · Catalogue: context 512000, reasoning, tools, modalities Text/Image
  • · First seen 2026-08-14

These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.

BES evidence selected · measured input only

The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.

Deployment readiness

Metadata index
77.1
strong · 94% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
100.0
Deployability
80.0
Evaluation provenance
37.9

Not yet published: the base model it derives from, pinned evaluation dataset versions.

A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.

Market activity

Venue coverage
Market Gravity
31.6
OpenRouter weekly popularity
#37 of 345

Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.

Arena preference

No matched Arena rating

This model is visible because it is available in the live catalogue, but it has no trustworthy community result yet. ModelCap does not infer a quality score from price, features, or market activity.

Market Gravity breakdown

Usage55%
38.2

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
22.9

Independent providers versus the model's open or closed cohort, weighted by uptime.

Open reach15%
9.3

Hugging Face 30-day downloads for open models; neutral for closed models.

Freshness5%
69.1

Time since first public availability, on a six-month half-life.

Specification

Model ID
dots-studio/dots-3-note-preview
Context window
512K tokens
Declared output cap
461K tokens

OpenRouter catalogue or provider-endpoint declaration. Limits can vary by provider and prompt length; a model-wide maximum has not been independently verified. OpenRouter metadata

Inputs
Text, Image
Outputs
Text
Tokenizer
Other
Cached input tokens
Not offered
First seen on OpenRouter
14 Aug 2026

Capabilities

  • Supported: Reasoning
  • Supported: Tool use
  • Supported: Structured output
  • Supported: Response format
  • Not supported: Moderated

Weights & access

Open weights. Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI.

Access
Open weights
Licence
apache-2.0
Pinned revision
1e1e7b0cd37a
Parameters
288.4B
Architecture
Dots3NoteForCausalLM
Model type
dots3_note
Hugging Face downloads (30d)
4K
Downloads (all time)
10K
Likes
295
Repository updated
15 Aug 2026, 08:00 UTC

Benchmark leaderboards featuring Dots3-Note Preview

All leaderboards
Public benchmark boards on which Dots3-Note Preview has a published result
LeaderboardScoreSource rankConfigurationPublished
ARC-AGI-2 leaderboardARC Prize Foundation76.8%#63 of 299Max18 Nov 2026

Alternatives to Dots3-Note Preview, by the numbers

Ranked neighbours

Models within 3 places of #50 on the ModelCap Index.

Dots3-Note Preview: common questions

How does Dots3-Note Preview rank among AI models?

Dots3-Note Preview holds ModelCap Index position #50 of 371 ranked language models as of 19 November 2026, with an Index score of 67.8 (interval 55.3–80.3); evidence: measured · blended. The Index combines public benchmark boards with published uncertainty; the methodology page explains the weighting.

How much does Dots3-Note Preview cost per 1M tokens?

Dots3-Note Preview is listed at Free per 1M input tokens and Free per 1M output tokens as of 19 November 2026. Prices come from the OpenRouter catalogue and re-observe on every refresh; at least one provider lists a free tier. For 1,000 requests using 2,000 input and 500 output tokens each, the listed rates imply Free total (2M input + 0.5M output). This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the chosen endpoint's rate and limits before budgeting.

What is the context window of Dots3-Note Preview?

Dots3-Note Preview has a published context window of 512K tokens (512,000). The snapshot lists a 460,800 token catalogue-declared route output cap. OpenRouter catalogue or provider-endpoint declaration. Limits can vary by provider and prompt length; a model-wide maximum has not been independently verified. The providers table lists each endpoint's context limit.

Is Dots3-Note Preview open-weight?

Open weights (apache-2.0). Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI. The repository is dots-studio/dots3-note-prev on Hugging Face.

Which API providers serve Dots3-Note Preview?

One provider serves Dots3-Note Preview through OpenRouter as of 19 November 2026: AtlasCloud. Each provider's price, context limit, quantization and measured uptime are in the providers table above.

Which benchmarks has Dots3-Note Preview been evaluated on?

Dots3-Note Preview has published results on 1 public board tracked by ModelCap as of 19 November 2026: ARC-AGI-2 76.8% (#63 of 299). Each board page ranks every tracked model on that benchmark. The ModelCap Index scores the results its methodology admits, with published uncertainty, and shows reference boards without scoring them.

When was Dots3-Note Preview released?

Dots3-Note Preview first appeared in the catalogue on 14 Aug 2026. It is the current version in its family. Rank and price movements since then are recorded on the site's changes feed.