Kimi Linear 48B A3B Instruct
Kimi Linear 48B A3B Instruct is Moonshot AI (Kimi)'s language model. A 48B-parameter MoE instruction-tuned LLM from Moonshot AI with 3B active parameters, designed for efficient linear-attention long-context reasoning.
Moonshot AI (Kimi)'s language model. A 48B-parameter MoE instruction-tuned LLM from Moonshot AI with 3B active parameters, designed for efficient linear-attention long-context reasoning.moonshot-kimi-48b-a3b-linear-instruct |
| Language |
| Active |
| Moonshot AI (Kimi) |
| Text |
| Text |
| 48B |
14.4#193 |
14.2#143 |
36.3#110 |
0.6#195 |
0.4#243 |
0.0#321 |
0.4#123 |
0.3#229 |
0.00s#101 |
0.2#226 |
0.4#110 |
0.3#128 |
0.1#101 |
0.0#264 |
0.0#307 |
Capabilities
Input1/5
✓
·
·
·
·
Output1/5
✓
·
·
·
·
Capabilities0/13
·
·
·
·
·
·
·
·
·
·
·
·
·
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| Kimi K2 Thinking Turbo | 262K | $1.15 | $8.00 | Available | |
| Kimi K2 Turbo | 256K | $1.15 | $8.00 | Available | |
| Kimi K2 Preview | — | 262K | $0.600 | $2.50 | Available |
| Kimi K2 Turbo Preview | — | 262K | $1.15 | $8.00 | Available |
| Kimi Linear 48B A3B Instruct | — | — | — | — | Current |
| Kimi | — | 131K | $2.00 | $5.00 | Available |
| Kimi 128K | — | 131K | $2.00 | $5.00 | Available |
| Kimi 32K | — | 33K | $1.00 | $3.00 | Available |
| Kimi 8K | — | 8K | $0.200 | $2.00 | Available |
| Kimi Thinking | — | 131K | — | — | Available |
| Kimi Thinking Preview | — | 131K | $0.600 | $2.50 | Available |