BenchLeader
InclusionAIOpen weightsReasoning modelAuto-detected

Ling 3.0 Tiny

Not yet ranked: too few independent results so far. Released 6 Aug 2026.

Blended price
Output speed
158 tok/s
First answer
15 s
Context
262k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Knowledge55
  2. Long context56
  3. Composite44

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarkdefaultSource
GPQA Diamond (AA)not in index73.4%#257Artificial Analysis
Humanity's Last Exam (AA)not in index9.3%#270Artificial Analysis

Coding

BenchmarkdefaultSource
SciCode (AA)not in index24.2%#154Artificial Analysis

Knowledge

BenchmarkdefaultSource
AA-Omniscience-19.4#181Artificial Analysis

Long context

BenchmarkdefaultSource
AA-LCR60.3%#216Artificial Analysis

Composite

BenchmarkdefaultSource
AA Intelligence Index11.9#274Artificial Analysis

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 30016.9 s
Summarise a 30-page report12,000 / 60018.8 s
Code edit6,000 / 1,50024.5 s
Agentic coding session60,000 / 4,00040.3 s
Structured extraction2,000 / 20016.3 s

See also

Data as of 9 Sept 2026. Compare with another model.