Ling 3.0 Flash

Official publisher releaseOpen weights

InclusionAI·#194 of 504 in the all-versions archive

Ling 3.0 Flash is a language model from InclusionAI. It ranks #194 of 504 in ModelCap's all-versions archive, with an Index score of 52.4, modeled from the publisher's launch results against measured models. Its score range of 35.0–69.8 spans positions #88–#315. It is listed at $0.021 per million input tokens and $0.063 per million output tokens across 2 providers. It reads up to 262K tokens of context, and its weights are openly downloadable.

Figures as of 6 Nov 2026, 13:52 UTC · download the public dataset · methodology

ModelCap Index · All-versions archive

52.4#194

Modeled from the publisher's launch results against measured models · range 35.0–69.8

Market Gravity

53.60.1

Secondary market signal

Superseded by Ling 3.1 flash 1019. It no longer holds a current-board position; its all-versions archive rank is #194.

Input

$0.021

per 1M tokens

Output

$0.063

per 1M tokens

Context

262K

tokens

Providers

endpoints

Providers

Uptime measured over the last 30 minutes

Provider	Input	Output	Context	Quant	Uptime
NovitaCheapest	$0.021	$0.063	262K	—	100.0%
DeepInfrabf16	$0.06	$0.18	131K	bf16	100.0%

Hugging Face Inference Providers

Exact repository identity

Provider economics and telemetry published by the Hugging Face Router. These are operational signals, not benchmark results.

Provider	Input	Output	Context	TTFT	Throughput	Features
deepinfra live	$0.06	$0.18	131K	279 ms	90.4 tok/s	tools · structured
novita live	$0.06	$0.18	262K	818 ms	241.4 tok/s	tools

Reported evaluation leads (5)

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 / task	Metric	Value	Revision	Provenance
cais/hle hle	—	22.7	Unpinned `not published`	Reported Not used in launch estimate
MathArena/aime_2026 MathArena/aime_2026	—	93.2	Unpinned `not published`	Reported Not used in launch estimate
MathArena/hmmt_feb_2026 MathArena/hmmt_feb_2026	—	87	Unpinned `not published`	Reported Not used in launch estimate
ScaleAI/SWE-bench_Pro SWE_Bench_Pro	—	56.6	Unpinned `not published`	Reported Not used in launch estimate
SWE-bench/SWE-bench_Multilingual swe_bench_multilingual_%_resolved	—	72.4	Unpinned `not published`	Reported Not used in launch estimate

Other versions (4)

ModelCap Index · All-versions archive

Methodology

52.4

Modeled from the publisher's launch results against measured models

All-versions archive

#194 of 504

Score range: 35.0–69.8
Rank range: #88–#315

Technical details

Public rank basis: Modeled · peer benchmarks (architecture)
Evidence label: Modeled · peer benchmarks · 4 reported rows against measured corpus ladders (67%); global-corpus-prior over 210 held-out anchors (33%); shrunk 0.5 up toward the measured corpus
Index support: 20.1% · limited
Identity binding: Exact catalogue product · inclusionai/inclusionai/ling-3.0-flash · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata inclusionAI/Ling-3.0-flash@fcc60a52b34c69c6271d0ac8a8cb0815020b6a43 (not the evaluation revision)
Index observations used: 0
Qualified Index evidence point: 52.4

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: 52.4 ± 13.6
Independent share: 0%
Shrink toward corpus: +0.5
Optimism haircut: None applied
Method: precision-weighted-v1

Channels fused into this model’s launch placement
Channel	Estimate	Share
Corpus ladder · not independent 4 reported rows against measured corpus ladders	51.9 ± 16.6	67%
Publisher corpus prior · not independent global-corpus-prior over 210 held-out anchors	53.5 ± 23.8	33%

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: —
Evidence status: No admitted measured evidence
General preference: —
Coding: —
Agents & tools: —
Reasoning: —
Evidence breadth: 0%
Evidence coverage: 0%

Market and catalogue signals · secondary, never the language Index rank

Usage (OpenRouter popularity): 26.4
Liquidity (providers × uptime): 99.0
Open reach (HF downloads): 73.5
Surface (context / tools / modalities): 75.3
Freshness: 66.8

· OpenRouter popularity #73
· 2 live providers
· HF 30d downloads 9548
· Catalogue: context 262144, reasoning, tools
· First seen 2026-07-23

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

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.

Experimental capability estimate

Estimate policy

49.1

35.2–62.9 uncertainty interval

Excluded from rank

Confidence: 79%
Training support: 221 models · 129 lineages
Out-of-domain check: in domain
Feature coverage: 100%
Lineage-held-out error: 8.3 MAE

Experimental estimate. This layer predicts from admitted evidence and safe metadata only. It never enters ModelCap Score or rank.

Deployment readiness

Metadata index

72.5

partial · 76% metadata coverage

Not capability

Artifact reproducibility: 85.0
Access & legal clarity: 100.0
Deployability: 90.0
Evaluation provenance: 0.0

Not yet published: the base model it derives from, admitted benchmark results, results across several capability areas, results from more than one source, a confident match to its benchmark entries, 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: 53.6
OpenRouter weekly popularity: #73 of 339

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%: 26.4
Liquidity25%: 99.0
Open reach15%: 73.5
Freshness5%: 66.8

Specification

Model ID: inclusionai/ling-3.0-flash
Context window: 262K tokens
Declared output cap: 33K 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
Outputs: Text
Tokenizer: Other
Cached input tokens: $0.0042 / 1M
First seen on OpenRouter: 23 Jul 2026

Capabilities

Supported: Reasoning
Supported: Tool use
Not 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: mit
Pinned revision: fcc60a52b34c
Parameters: 127.5B
Architecture: BailingMoeV3ForCausalLM
Model type: bailing_hybrid
Hugging Face downloads (30d): 10K
Downloads (all time): 49K
Likes: 437
Repository: inclusionAI/Ling-3.0-flash
Repository updated: 12 Oct 2026, 11:46 UTC

Ling 3.0 Flash: common questions

How does Ling 3.0 Flash rank among AI models?

Ling 3.0 Flash is a superseded version and is not on the current board as of 6 November 2026. Its archive position, where one exists, is shown on the page; the newer version carries the current rank.

How much does Ling 3.0 Flash cost per 1M tokens?

Ling 3.0 Flash is listed at $0.021 per 1M input tokens and $0.063 per 1M output tokens as of 6 November 2026 across 2 providers. Prices come from the OpenRouter catalogue and re-observe on every refresh. For 1,000 requests using 2,000 input and 500 output tokens each, the listed rates imply $0.073 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 Ling 3.0 Flash?

Ling 3.0 Flash has a published context window of 262K tokens (262,144). The snapshot lists a 32,768 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 Ling 3.0 Flash open-weight?

Open weights (mit). 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 inclusionAI/Ling-3.0-flash on Hugging Face.

Which API providers serve Ling 3.0 Flash?

2 providers serve Ling 3.0 Flash through OpenRouter as of 6 November 2026: Novita and DeepInfra. Each provider's price, context limit, quantization and measured uptime are in the providers table above.

When was Ling 3.0 Flash released?

Ling 3.0 Flash first appeared in the catalogue on 23 Jul 2026. It has since been superseded by Ling 3.1 flash 1019. Rank and price movements since then are recorded on the site's changes feed.