Updated 4 h ago · first seen 11 Sept 2026
model_01M29DV4DKDBPPS82CHKF70QR4
- Parameters
- 218B · 25B active
- T2 · 4 h ago
- Context
- 16K tokens
- T2 · 4 h ago
- Released
- 10 Sept 2026
- T2 · 4 h ago
- License
- CC BY-NC 4.0
- T2 · 4 h ago
Specification
- Family
- North
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Version
- 1.0
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Release date
- 10 Sept 2026
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Status
- announced
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Openness
- open-weights
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- License
- CC BY-NC 4.0
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Architecture
- MoE
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Parameters
- 218B
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Active parameters
- 25B
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Mixture of experts
- Yes
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Context window
- 16K tokens
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Max output
- 16K tokens
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Modalities
- Text
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Input modalities
- Text
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Output modalities
- Text
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Languages
- 50+ languages
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
17
Source tiers
T217
Freshest observation
4 h ago
Conflicts
None
Modalities
- Modalities
- Text
- Input
- Text
- Output
- Text
Capabilities
Tool calling
Unavailable
Structured output
Unavailable
Reasoning
Unavailable
Vision
Unavailable
Audio
Unavailable
Fine-tuning available
Unavailable
No capability flags have been observed from a source yet — we do not infer them.
- Context window
- 16K tokens
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Max output
- 16K tokens
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
- Languages
- 50+ languages
Source:Cohere — docs & blogT2observed 4 h agomediumLLM-extracted
No benchmark results recorded
Current prices
No current prices recorded
Price history
Memory need = bytes per parameter (4-bit ≈ 0.5 × 1.15 overhead, 8-bit 1.0, fp16 2.0) + a KV-cache allowance. Not a measurement.
| Hardware | Quantization | Memory | Est. need | Fits |
|---|---|---|---|---|
| NVIDIA DGX B200 | 4bit | 1,440 GB | 125.8 GB | Yes |
| Apple M3 Ultra | 4bit | — | 125.8 GB | Yes |
| Mac Studio (Apple M5 Ultra) | 4bit | — | 125.8 GB | Yes |
| AMD Instinct MI325X | 4bit | 256 GB | 125.8 GB | Yes |
| AMD Instinct MI300X | 4bit | 192 GB | 125.8 GB | Yes |
| Apple M2 Ultra | 4bit | — | 125.8 GB | Yes |
| NVIDIA B200 | 4bit | 180 GB | 125.8 GB | Yes |
| NVIDIA H200 | 4bit | 141 GB | 125.8 GB | Yes |
| NVIDIA H200 NVL | 4bit | 141 GB | 125.8 GB | Yes |
| Apple M1 Ultra | 4bit | — | 125.8 GB | Yes |
| Apple M3 Max | 4bit | — | 125.8 GB | Yes |
| Apple M4 Max | 4bit | — | 125.8 GB | Yes |
| Mac Studio (Apple M5 Max) | 4bit | — | 125.8 GB | Yes |
| MacBook Pro (Apple M5 Max) | 4bit | — | 125.8 GB | Yes |
| NVIDIA DGX Spark | 4bit | 128 GB | 125.8 GB | Yes |
| Apple M2 Max | 4bit | — | 125.8 GB | No |
| NVIDIA H100 NVL | 4bit | 94 GB | 125.8 GB | No |
| NVIDIA A100 80GB | 4bit | 80 GB | 125.8 GB | No |
| NVIDIA H100 SXM | 4bit | 80 GB | 125.8 GB | No |
| Apple M1 Max | 4bit | — | 125.8 GB | No |
| Apple M4 Pro | 4bit | — | 125.8 GB | No |
| Mac mini (Apple M5 Pro) | 4bit | — | 125.8 GB | No |
| MacBook Pro (Apple M5 Pro) | 4bit | — | 125.8 GB | No |
| Apple M3 Pro | 4bit | — | 125.8 GB | No |
| Apple M1 Pro | 4bit | — | 125.8 GB | No |
| Apple M2 Pro | 4bit | — | 125.8 GB | No |
| Apple M4 | 4bit | — | 125.8 GB | No |
| iMac (Apple M4) | 4bit | — | 125.8 GB | No |
| Mac mini (Apple M6) | 4bit | — | 125.8 GB | No |
| MacBook Air (Apple M5) | 4bit | — | 125.8 GB | No |
| MacBook Pro (Apple M5) | 4bit | — | 125.8 GB | No |
| NVIDIA GeForce RTX 5090 | 4bit | 32 GB | 125.8 GB | No |
| Apple M2 | 4bit | — | 125.8 GB | No |
| Apple M3 | 4bit | — | 125.8 GB | No |
| NVIDIA GeForce RTX 3090 | 4bit | 24 GB | 125.8 GB | No |
| NVIDIA GeForce RTX 4090 | 4bit | 24 GB | 125.8 GB | No |
| Apple M1 | 4bit | — | 125.8 GB | No |
No lineage recorded
Papers 0
No papers linked yet.
Repositories 0
No repositories linked yet.
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Architecture
Architecturearchitecture1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| MoE | → current | current | Cohere — docs & blogT2 | medium | llm |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Cohere — docs & blog | cohere.com/blog/north-small-translate | news | T1· Official | 3 h ago | 5 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.