BenchLeader
MetaAuto-detected

Llama 2-7B

Not yet ranked: too few independent results so far.

Blended price
$0.100/M
$0.050 in · $0.250 out
Output speed
First answer
Context
4k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Reasoning22
  2. Coding19
  3. Human preference23
  4. Composite36

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 Coding1103#333LMArena

Human preference

BenchmarkdefaultSource
LMArena Text1107#334LMArena

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$0.0001
Summarise a 30-page report12,000 / 600$0.0008
Code edit6,000 / 1,500$0.0007
Agentic coding session60,000 / 4,000$0.0040
Structured extraction2,000 / 200$0.0001

See also

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