BenchLeader
Ant-groupAuto-detected

Ling Flash 2.0

Best configuration ranks #400 of 610 on the BenchLeader Index at 45.0 ±5.5. Last measured 2 Sept 2026. Released 18 Sept 2025.

Blended price
$0.248/M
$0.140 in · $0.570 out
Output speed
7 tok/s
First answer
2.68 s
first token 2.68 s
Context
131k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index45
  2. Reasoning52
  3. Coding54
  4. Agents & tools43
  5. Knowledge34
  6. Instruction following38
  7. Human preference52
  8. Long context34
  9. Composite39

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

Coding

BenchmarkdefaultSource
LMArena Coding1411#165LMArena

Agents & tools

Knowledge

BenchmarkdefaultSource
AA-Omniscience-62.7#407Artificial Analysis

Instruction following

BenchmarkdefaultSource
IFBench34.4%#304Artificial Analysis

Human preference

BenchmarkdefaultSource
LMArena Text1344#186LMArena

Long context

BenchmarkdefaultSource
AA-LCR17.7%#372Artificial Analysis

Composite

BenchmarkdefaultSource
AA Intelligence Index7.8#380Artificial 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 / 300$0.000243.7 s
Summarise a 30-page report12,000 / 600$0.00201.4 min
Code edit6,000 / 1,500$0.00173.5 min
Agentic coding session60,000 / 4,000$0.0119.2 min
Structured extraction2,000 / 200$0.000430.0 s

See also

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