cohere
Cohere: North Mini Code (free)
cohere/north-mini-code:free
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
cohere/north-mini-code:free is the free OpenRouter listing of Cohere's North Mini Code, an open-weights research release by Cohere and Cohere Labs under Apache 2.0. The model card describes a sparse mixture-of-experts design with 30B total and 3B active parameters, text-only input and output, a 256K context window with up to 64K output, and post-training aimed at agentic coding. Vendor-reported capabilities include trained tool use, interleaved thinking, and terminal-oriented tasks. The OpenRouter catalog additionally reports Artificial Analysis index measurements. Independent evaluation remains limited.[1][2]
Strengths
- Vendor-described agentic coding focus: post-training combined cascaded supervised fine-tuning with reinforcement learning from verifiable rewards, plus tool use trained through chat templates and JSON schema descriptions.[2]
- Large context for a low active-parameter count: the card reports a 256K context window and 64K maximum output, with only 3B of 30B parameters active per token.[1][2]
- Apache 2.0 open weights runnable locally via transformers or vLLM per the card, and listed as a free variant in the OpenRouter catalog.[1][2]
Limitations
- Text-only modality: no image, audio, or other input or output types are reported by either source.[1][2]
- Evidence gaps: benchmark figures in the card appear only inside an image, and apart from Artificial Analysis indices in the OpenRouter catalog, evaluations are vendor-reported; knowledge cutoff is not stated.[1][2]
- Deployment is not turnkey: the card calls for building vLLM main with Cohere's melody library and installing transformers from source.[2]
Where to use it
- Editorial suggestion: agentic coding assistants that invoke tools such as a shell, given the card's tool-use training and JSON schema guidance.[2]
- Terminal and software-engineering workflows where long context may help; validate on your own workload first.[2]
- No-cost experimentation through the free OpenRouter listing before self-hosting the open weights.[1][2]
Sources
- [1] OpenRouter catalog · Cohere: North Mini Code (free)Sources checked: Nov 13, 2026
- [2] Model documentation · Cohere: North Mini Code (free)Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
North Mini Code is Cohere's first agentic coding model and the debut of its North family. A sparse mixture-of-experts model with 30B total parameters and 3B active, it is optimized...
Specifications
- Model ID
- cohere/north-mini-code:free
- Canonical ID
- cohere/north-mini-code-20260617
- Creator
- cohere
- Added to OpenRouter
- 17.06.2026
- Context length
- 256,000
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0 / out $0
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | CC — below the B band | reasoning: optional · parameters: include_reasoning, reasoning |
| Coding | CC — below the B band | — |
| Agents | CC — below the B band | parameters: tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Intelligence index | 3.4 | CC — below the B band | 13.11.2026 |
| Artificial Analysis | Coding index | 36.5 | CC — below the B band | 13.11.2026 |
| Artificial Analysis | Agentic index | 1.1 | CC — below the B band | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| Cohere | 256,000 | $0 | $0 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.