North Mini Code
Jun 2026
North Mini CodeCompare inon
Benchmarks of North Mini Code19
BenchmarkByScoreRank
Overall1
Knowledge4
Science3
Coding3
Agents4
Documents2
Legacy · saturated or no longer reported2
No benchmark matches
—per task?
free API onlyprice
81 tok/sfirst token in 0.5 s?
256Kcontext
Ranking?
Alternatives?
ModelScoreCostPer taskWhat changes
BetterGPT-6 Lunalow×2.2—$0.005Function calling only with none · Chat Completions+3
OpenQwen3.8 27Bxhigh×3.4—$1.01max_tokens range · QwenCloud+4BetterGPT-5.6 Lunalow×2.1—$0.01
BetterMiMo v2.5thinkingOpen×2.2—$0.019Memory (4-bit): 18 GB → 186 GB
BetterClaude Haiku 5.5low×3—$0.024Remove sampling parameters+5ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
GPT-6 Lunalow×2.2—$0.005Function calling only with none · Chat Completions+3GPT-5.6 Lunalow×2.1—$0.01MiMo v2.5thinkingOpen×2.2—$0.019Memory (4-bit): 18 GB → 186 GBClaude Haiku 5.5low×3—$0.024Remove sampling parameters+5GPT-5 Minihigh+69%—$0.053ModelScoreCostPer taskWhat changes
Qwen3.8 27BxhighOpen×3.4—$1.01max_tokens range · QwenCloud+4IK2 Horizon MoVA 36B A4BthinkingOpen×2.6—$1 / $4Memory (4-bit): 18 GB → 22 GBQwen 3.5 27BthinkingOpen×2.3—$0.30 / $2.40AG9v3 39A5BOpen×2.2——Context: 256K → 131K+1Qwen3.6 27BthinkingOpen×2.2—$0.62ModelScoreCostPer taskWhat changes
Command A+thinkingOpen+32%—$0.30 / $1.50Context: 256K → 128K+1Command Anon-thinkingOpen-30%—$2.50 / $10Memory (4-bit): 18 GB → 67 GBTiny Aya GlobalOpen-51%——Context: 256K → 8K+1Command R7B (12-2024)Openn/a—$0.037 / $0.15Context: 256K → 128K+1Command R (08-2024)unstatedOpenn/a—$0.15 / $0.60Context: 256K → 128K+1Command R+ (08-2024)unstatedOpenn/a—$2.50 / $10Context: 256K → 128K+1Command R+Openn/a——Context: 256K → 128K+1Command ROpenn/a——Context: 256K → 128K+1All Cohere models →Run it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✓ 24 GB GPU · compressed✓ 48 GB GPU · 8-bit✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
Benchmarks
all categoriesChat & Knowledge
Humanity's Last Exam11.1#173
AA Omniscience Index-48.6#203
CritPt (physics)0.3#149
AA Intelligence Index 4.3.29.9#209
AA Omniscience · Accuracy18.9#167
AA-LCR (long context) 1.137.3#206
GDP.pdf · AA all-pass1.8#107
GPQA Diamond75.7#149
Coding
Terminal-Bench 4.00.5#63
SciCode38.8#102
Terminal-Bench 2.135.6#91
Agents & Tool Use
AutomationBench-AA1.5#105
τ³-Banking · AA6.4#107
AA-Briefcase 1.1 · Elo211.8#116
τ²-bench · AA37.4#131
Writing & Language
IFBench (instructions)57.6#93
FAQ
How much does North Mini Code cost?›
There is no per-unit API price: free API only.
Is North Mini Code free to use?›
Yes, there is a free API option.
How good is North Mini Code, and what is it best at?›
#172 of 334 in Chat & Knowledge, #63 of 132 in Coding, #105 of 133 in Agents & Tool Use, #93 of 245 in Writing & Language, #107 of 133 in Vision & Documents.
Is North Mini Code better than GPT OSS 20B?›
On AA Intelligence Index 4.3.2, North Mini Code scores 9.9 vs 10 — about the same.
What is the context window of North Mini Code?›
256K tokens, with up to 64K output tokens.
When was North Mini Code released? Is there a newer version?›
Released 2026-06-09. It is the current version in its line.
Is North Mini Code open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run North Mini Code locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can North Mini Code read images?›
No, it accepts text only.
What is North Mini Code?›
North Mini Code is an open-weights language model from Cohere, released Jun 2026.
What are the best alternatives to North Mini Code?›
GPT-6 Luna (better), Qwen3.8 27B (open weights).