AI Infra Interviews logo

Ling / Hybrid KDA / gated MLA MoE

Ling-3.1-flash-1019

A post-training update needs its own deployment evidence

Checkpoint uploadSources checked 2026-11-16

Evaluate this revision against the original on identical tools, prompts and reasoning budgets. The model card reports post-training gains; unchanged geometry does not establish unchanged behavior.

560Breported total parameters
25Bactive parameters per token
262,144configured token positions

Publisher-reported rounded count; the pinned tensor index is a separate payload measurement.

Evaluate it for

Text-agent evaluation with a runtime that supports this exact hybrid layout.

Choose another path when

A workstation budget, image inputs, or a requirement for already-qualified million-token concurrency.

Release context

Why this release matters

The two releases share 560B total and 25B active counts. Their artifact identity, task results and rollout acceptance still need separate records.

The repository records the weight upload on this date. The 1019 suffix is not proof of an October 19 public weight release.

Release evidence · Follow the Ling timeline →

Deployment starting points

Which hardware, how many, at what precision?

Start with an exact artifact and an engine that supports it. A configuration below is evidence of a documented path; its device count does not promise a particular throughput or concurrency.

No certified hardware configuration in this edition

A workstation budget, image inputs, or a requirement for already-qualified million-token concurrency.

The pinned model card establishes the architecture. It does not, by itself, prove that a particular GPU count runs this checkpoint at your target context. The capacity screen below helps rule out undersized allocations; a successful load and workload test are still required.

573.75 GB of tensor data in the pinned index

This is the index’s reported tensor payload, in decimal GB. It excludes file headers and runtime memory. Mixed precision, conversion and host/device placement determine how much GPU memory the loaded model needs.

Inspect the exact index metadata →
Capacity screen: would the weights fit?

This estimates inference memory for a hypothetical uniform precision. It is useful for rejecting an allocation that is too small. It does not establish a working deployment or estimate training memory.

A uniform-precision GPU count would hide this model’s component layout. Publisher-reported rounded count; the pinned tensor index is a separate payload measurement. Use the documented artifact and placement path above; no GPU count is inferred here.

Architecture in practice

What the serving engine has to do

Two kinds of layers retain two kinds of state.

A token enters a layerthe layer type selects itspathRecurrent layerupdate a compact stateAttention layerread retained tokenhistoryRecurrent statemaintained for thissequenceAttention cachelayout depends on themodel

Hybrid attention does not mean zero request state.

A full-attention formula cannot describe every layer.

Schematic of a hybrid attention stack, not a literal layer ordering. Recurrent layers update a compact state; attention layers retain their own history. Both kinds of state matter for prefix reuse and concurrent requests.

Mechanism schematic based on the pinned model card and configuration. It explains a design principle; it is not a full implementation graph.

Inspect the full checkpoint specifications
Inspected fields for this exact revision. KV heads alone do not describe latent, hybrid or shared-cache layouts.
Exact checkpointinclusionAI/Ling-3.1-flash-1019
Text model typebailing_hybrid
Layers72
Hidden width4,096
Attention / KV heads64 / 64
Head dimensionNot verified
Routed / selected experts512 / 8
Configured positions262,144
Documented extensionNot recorded for this checkpoint
Inputs → outputtext → text
License metadataapache-2.0

63 KDA layers; 9 gated MLA layers. These counts describe the configured pattern.

Open weights do not imply unrestricted use. Read the applicable license terms linked from the pinned card.

A useful comparison

What else belongs on the shortlist?

Compare the two Ling revisions under one harness. The smaller alternative changes residency costs as well as possible task quality.

13 Oct 2026

Ling-3.1-flash

Count recurrent and latent state separately

560B reported total · text

14 Aug 2026

Qwen3.8-27B

A current dense baseline before you scale out

27B reported total · text + image + video

These are editorial comparison candidates. We have not run a matched quality or serving benchmark, so this is not a ranking.

Prepare to explain it

The interview lesson

Explain why a copied launch command and a date-like model suffix are insufficient deployment evidence.

Attention layouts · Quantization · Courses and worked examples

For a deployment evaluation, record exact weights, precision, engine, device count, interconnect, prompt/output lengths and concurrency. Report task success, errors, TTFT, TPOT and useful throughput together.

Sources and citation

Our assessment is editorial judgment based on the linked architecture and deployment evidence. Model facts are publisher-reported or attributed to runtime maintainers; no independent GPU benchmark was run.

AI Infra Interviews, “Ling-3.1-flash-1019: release, architecture and deployment”, checked 2026-11-16. Preserve this date and the exact checkpoint when citing.

Model and research notes as JSON · InclusionAI profile