Skip to content

Ling 3.1 Flash

InclusionAIOpen weightsReasoningProvisionalReleased Sep 29, 2026Input: text
Compare with GLM-5.3Source scores ↗OpenRouter ↗
Overall score
68.1
Rank 24 of 222 · 53.7–82.4
Context window
262K
262,000 tokens
Price per 1M tokens
$0.075 in · $0.23 out
Source: OpenRouter
Blended price
$0.11
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Agentic & tool use82.3
  2. 2Reasoning82.0
  3. 3Coding69.1
  4. 4Knowledge57.4
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (15)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

Agentic & tool use

Terminal-Bench 2.181.2#27 of 68
BrowseComp91.7#3 of 50
MCP Atlas83.0#8 of 42
AutomationBench52.5#5 of 25
CyberGym87.9#4 of 25
τ³-bench results47.8#15 of 23
Terminal-Bench 4.046.03#9 of 18
WideResearch83.3#2 of 16
DRACO85.5#3 of 6

Coding

SWE-bench Pro68.7#8 of 69
DeepSWE59.7#36 of 50
SWE-Atlas Codebase QnA64.5#1 of 4

Knowledge

HLE37.4#33 of 69
HealthBench Professional65.3#3 of 16

Multilingual

SWE Multilingual88.5#6 of 46

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
List price (OpenRouter)$0.075$0.23—
OpenRouter inclusionai/ling-3.1-flash$0.075$0.23$0.015

MIT-licensed FP8 weights are now available for self-hosting. InclusionAI does not publish an exact first-party hosted USD token rate for this checkpoint, so canonical hosted prices remain unresolved. Vercel’s separate Novita route is paid after its October 13, 2026 promotion; consult the linked Gateway listing for that route’s rates and 262,144-token context limit. Self-hosting requires hardware and has no provider token bill.

Open weights

From the Hugging Face repository.

Parameters
572B
Licence
mit
Downloads (30 days)
2.9K

Speed

Measured by us through OpenRouter (median of trials).

Not measured yet.

Related models

Same family

Similar score