Skip to content
ModelCap

Qwen3.8 Flash

Current board #17Official publisher releaseRestricted license7d #108 → #17

Qwen·qwen/qwen3.8-flash

Qwen3.8 Flash is ModelCap current-board rank #17 with Index 72.1 (Measured), from the ModelCap Index as of 3 Sept 2026, 02:09 UTC.

Qwen3.8 Flash listed OpenRouter input price is $0.15 per 1M tokens and listed output price is $0.47 per 1M tokens, from OpenRouter listed prices as of 3 Sept 2026, 02:09 UTC.

Qwen3.8 Flash has a published context limit of 1M tokens in the ModelCap catalogue as of 3 Sept 2026, 02:09 UTC.

Qwen3.8 Flash has 1 published OpenRouter serving endpoint, from OpenRouter as of 3 Sept 2026, 02:09 UTC.

Qwen3.8 Flash weight access is restricted license, from ModelCap classification of published repository metadata as of 3 Sept 2026, 02:09 UTC.

Snapshot facts · download the public dataset · methodology

ModelCap Index · Current board
72.1#17
Measured · 20% support · score interval 72.1–88.9
Evidence as of 2 Sept 2026, 20:11 UTC
Market Gravity
37.6Increased by 0.8
Secondary market signal
Input
$0.15
per 1M tokens
Output
$0.47
per 1M tokens
Context
1M
tokens
Providers
1
endpoint
Gateway spend
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
Alibaba$0.15$0.471M100.0%

Reported evaluation leads (11)

Excluded from score

Structured rows reported in this exact model repository at the pinned revision below. They are useful research leads, but remain quarantined until dataset revision, harness, configuration and identity pass ModelCap’s independent admission review.

Dataset / taskMetricValueRevisionProvenance
cais/hle
hle
35.9
Unpinned
not published
Reported
Quarantined
claw-eval/Claw-Eval
multimodal
64.4
Unpinned
not published
Reported
Quarantined
datacurve/deep-swe
deep_swe
58.7
Unpinned
not published
Reported
Quarantined
hkust-nlp/Toolathlon
toolathlon_verified
73.5
Unpinned
not published
Reported
Quarantined
Idavidrein/gpqa
diamond
91.7
Unpinned
not published
Reported
Quarantined
llamaindex/ExtractBench
mean
89.88
Unpinned
not published
Reported
Quarantined
llamaindex/ExtractBench
short
94.82
Unpinned
not published
Reported
Quarantined
llamaindex/ExtractBench
medium
87.81
Unpinned
not published
Reported
Quarantined
llamaindex/ExtractBench
long
37.74
Unpinned
not published
Reported
Quarantined
ScaleAI/SWE-bench_Pro
SWE_Bench_Pro
62.5
Unpinned
not published
Reported
Quarantined
SWE-bench/SWE-bench_Multilingual
swe_bench_multilingual_%_resolved
81
Unpinned
not published
Reported
Quarantined

Other versions (3)

ModelCap Index · Current board

Methodology
72.1
Measured · 1 public benchmark observation across 1 board
Current board
#17
Public rank basis
Measured (measured)
Index support
20.0% · limited
Index interval
72.1–88.9
Rank posterior
Not available
Top 5 / Top 10 probability
Not available / Not available
Posterior as of
Snapshot timestamp unavailable
Evidence as of
2 Sept 2026, 20:11 UTC
Identity binding
Exact catalogue product · qwen/qwen/qwen3.8-flash · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata Qwen/Qwen3.8-Flash-Next@de4b8e4d43b917e7706784d8bb445c9af86a3540 (not the evaluation revision)
Index observations used
1
Qualified Index evidence point
80.5
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
80.5
Evidence status
provisional · 50% mass
General preference
80.5
Coding
Agents & tools
Reasoning
Evidence breadth
80%
Evidence coverage
20%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
30.1
Liquidity (providers × uptime)
34.5
Open reach (HF downloads)
Surface (context / tools / modalities)
100.0
Freshness
97.3
  • · OpenRouter popularity #55
  • · 1 live provider
  • · Catalogue: context 1000000, reasoning, tools, modalities Text/Image/Video
  • · First seen 2026-08-26

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
75.6
strong · 94% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
92.5
Deployability
80.0
Evaluation provenance
37.9

Missing: artifactReproducibility.base-lineage, evaluationProvenance.evaluation-dataset-revisions.

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

Market activity

Venue coverage
Market Gravity
37.6
OpenRouter weekly popularity
#55 of 310

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%
30.1

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
34.5

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

Open reach15%
50.0

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

Freshness5%
97.3

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

Specification

Context window
1M tokens
Max output
131K tokens
Inputs
Text, Image, Video
Outputs
Text
Tokenizer
Qwen
Cached input tokens
$0.016 / 1M
First seen on OpenRouter
26 Aug 2026

Capabilities

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

Weights & access

Restricted license. Downloadable weights use a model-specific or use-restricted license. Review the linked license before use, redistribution, or commercial deployment.

Access
Restricted license
Pinned revision
de4b8e4d43b9
Parameters
180.0B
Architecture
Qwen4ExpForConditionalGeneration
Model type
qwen4_exp
Hugging Face downloads (30d)
208K
Downloads (all time)
208K
Likes
5K
Repository updated
27 Aug 2026, 05:03 UTC

Compare Qwen3.8 Flash with

Pick any model

Benchmark leaderboards featuring Qwen3.8 Flash

All leaderboards
Public benchmark boards on which Qwen3.8 Flash has a published result
LeaderboardScoreSource rankConfigurationPublished
Artificial Analysis Intelligence IndexArtificial Analysis55.8#31 of 623intelligence-index2 Sept 2026
Terminal-Bench 2.1 (Artificial Analysis run)Artificial Analysis86.1%#15 of 226terminal-bench-2.12 Sept 2026
Humanity's Last Exam (Artificial Analysis run)Artificial Analysis38.0%#58 of 591humanitys-last-exam2 Sept 2026

Alternatives to Qwen3.8 Flash, by the numbers

Ranked neighbours

Models within 3 places of #17 on the ModelCap Index.

Other ranked Qwen models

The same publisher's other current models with a public Index position.

Qwen3.8 Flash: common questions

How does Qwen3.8 Flash rank among AI models?

Qwen3.8 Flash holds ModelCap Index position #17 of 225 ranked language models as of 3 September 2026, with an Index score of 72.1 (interval 72.1–88.9); evidence: measured. The Index combines public benchmark boards with published uncertainty; the methodology page explains the weighting.

How much does Qwen3.8 Flash cost per 1M tokens?

Qwen3.8 Flash is listed at $0.15 per 1M input tokens and $0.47 per 1M output tokens as of 3 September 2026. Prices come from the OpenRouter catalogue and re-observe on every refresh.

What is the context window of Qwen3.8 Flash?

Qwen3.8 Flash has a published context window of 1M tokens (1,000,000) and a maximum output of 131K tokens. Individual providers can serve less than the published maximum; the providers table lists each endpoint's own limit.

Is Qwen3.8 Flash open-weight?

Restricted license (qwen-community-1.0). Downloadable weights use a model-specific or use-restricted license. Review the linked license before use, redistribution, or commercial deployment. The repository is Qwen/Qwen3.8-Flash-Next on Hugging Face.

Which API providers serve Qwen3.8 Flash?

One provider serves Qwen3.8 Flash through OpenRouter as of 3 September 2026: Alibaba. Each provider's price, context limit, quantization and measured uptime are in the providers table above.

Which benchmarks has Qwen3.8 Flash been evaluated on?

Qwen3.8 Flash has published results on 3 public boards tracked by ModelCap as of 3 September 2026: AA Intelligence Index 55.8 (#31 of 623), AA Terminal-Bench 2.1 86.1% (#15 of 226) and AA Humanity's Last Exam 38.0% (#58 of 591). Each board page ranks every tracked model on that benchmark; the ModelCap Index combines them with published uncertainty.

How does Qwen3.8 Flash compare with GPT-5.6 Luna?

Qwen3.8 Flash ranks #17 (Index 72.1) and GPT-5.6 Luna ranks #18 (Index 69.7) on the ModelCap Index as of 3 September 2026. The head-to-head page lines up their benchmarks, listed prices, context windows and provider counts side by side.

When was Qwen3.8 Flash released?

Qwen3.8 Flash first appeared in the catalogue on 26 Aug 2026. It is the current version in its family. Rank and price movements since then are recorded on the site's changes feed.