Ling-2.6 1T

Language Ant Group Partly verified Released Apr 29, 2026

Ling-2.6-1T is an open-source model in the Ling family for coding and daily workflows. Its API supports OpenAI and Anthropic protocols, while the release entry for Ling-2.6-flash specifies OpenAI protocol access. Summary written by AI Index from Ant Group’s own pages.

  • Out Output modalities: Text
  • Reasoning
  • Tool use
  • Open weights

Prices

1 offering · prices as listed
Ant Ling API (opens in a new tab) Official China · CNY Ling-2.6-1T
Conditions/ 1M tokensInputCache readOutput
Standard no longer listed since Oct 20, 2026 ¥4.50¥0.90¥18.00

Prices from Ant Ling: model pricing (opens in a new tab) (checked Oct 19, 2026)

Standard: the price a request with no conditions pays (the headline price elsewhere on the site).

Details

Family
ling
Model ID

Not stated on Ant Group’s pages: max input, input modalities, reasoning efforts, reasoning can be off, max thinking budget, structured output, built-in tools, knowledge cutoff, licence, aliases.

Run locally

Weights: inclusionAI/Ling-2.6-1T
  • Parameters 1.03T
  • Active 76.8B (estimated) · 256 experts, 8 per token
  • Native precision BF16
  • Layers 80
  • KV cache per token 90 KB
  • Config positions 256K (config.json max_position_embeddings)
  • API context 977K (maker’s API)
My hardware GeForce RTX 4090 24 GB · 1008 GB/s · 64 GB DDR5 dual channel · 89.6 GB/s
On your hardware · GeForce RTX 4090
Doesn’t fit IQ4_XS (automatic) 8K context
WeightsKV cacheOverheadUsable
Needs
≈ 533.4 GB (weights 507.3 + KV cache 0.7 + overhead 25.4)
Usable
21.6 GB (90% of its memory)
Try
it’s too large for this hardware; see the smallest that runs it below
PrecisionWeightsNeeds at 8KOn your hardwareLongest context
BF16 (native)maker’s file970.3 GB≈ 1019.5Doesn’t fit—
Q8_01014.7 GB≈ 1066.1Doesn’t fit—
FP8955.0 GB≈ 1003.4Doesn’t fit—
Q6_K784.3 GB≈ 824.2Doesn’t fit—
Q5_K_M679.2 GB≈ 713.9Doesn’t fit—
Q5_0656.6 GB≈ 690.1Doesn’t fit—
Q4_K_M580.2 GB≈ 609.9Doesn’t fit—
Q4_0537.2 GB≈ 564.7Doesn’t fit—
IQ4_XS507.3 GB≈ 533.4Doesn’t fit—
Q3_K_Mquality drops466.8 GB≈ 490.8Doesn’t fit—
IQ3_XXSquality drops365.3 GB≈ 384.3Doesn’t fit—
Q2_Kquality drops358.1 GB≈ 376.7Doesn’t fit—
IQ2_XXSquality drops245.9 GB≈ 258.9Doesn’t fit—

Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.

Compared with other hardware

Each at the highest precision it runs. Click one for details.

Smallest that runs it (Q4_K_M, 8K): no single listed machine; about 9 × 80 GB GPUs.

Run commands

The maker publishes no files at this precision. Choose its native precision or one of its own builds, or quantise it yourself with llama.cpp.

Memory to fine-tune
Full≈ 15283.4 GBDoesn’t fit your hardware
LoRA≈ 1914.8 GBDoesn’t fit your hardware
QLoRA≈ 542.0 GBDoesn’t fit your hardware

Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.

Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-21. How it’s estimated

Cited scores

From third-party leaderboards, not measured by us

This model isn’t on any leaderboard we cite yet.

No changes recorded yet.

Sources

Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data

Ant Group · China · developer.ant-ling.com

Snapshots, the same family and other models from this provider