CohereOpen weightsReasoning modelAuto-detected

North Mini Code

Name: North Mini Code benchmark results
Creator: BenchLeader

Best configuration ranks #301 of 610 on the BenchLeader Index at 49.8 ±6.9. Released 9 Jun 2026.

Blended price: –
Output speed: 44 tok/s
First answer: 46 s
Context: 256k

How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.

Overall index50
Agents & tools61
Knowledge41
Instruction following58
Long context44
Composite45

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

Benchmark	default	Source
GPQA Diamond (AA)not in index	75.7%#238	Artificial Analysis
Humanity's Last Exam (AA)not in index	11.1%#237	Artificial Analysis

Coding

Benchmark	default	Source
SciCode (AA)not in index	38.8%#122	Artificial Analysis

Agents & tools

Benchmark	default	Source
Terminal-Bench Hard	31.1%#104	Artificial Analysis
τ²-Bench Telecom (AA)not in index	37.4%#220	Artificial Analysis

Knowledge

Benchmark	default	Source
AA-Omniscience	-48.6#305	Artificial Analysis

Instruction following

Benchmark	default	Source
IFBench	57.5%#135	Artificial Analysis

Long context

Benchmark	default	Source
AA-LCR	37.3%#297	Artificial Analysis

Composite

Benchmark	default	Source
AA Intelligence Index	12.8#255	Artificial 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.

Workload	Tokens in / out	Cost	With caching	Time
Chat reply	400 / 300	–	–	53.0 s
Summarise a 30-page report	12,000 / 600	–	–	59.8 s
Code edit	6,000 / 1,500	–	–	1.3 min
Agentic coding session	60,000 / 4,000	–	–	2.3 min
Structured extraction	2,000 / 200	–	–	50.7 s