- Intelligence
- 8.5
- Rank 488 of 744 measured models
- Coding
- 23.0
- Rank 211 of 298 measured models
- Output speed
- 37 t/s
- Rank 374 of 403 measured models
- Price/1M
- $0.463
- Rank 268 of 659 measured models
- Context window
- 128K
- Rank 773 of 1199 measured models
- Cost per task
- $0.0198
- Rank 23 of 193 measured models
DeepSeek V3 (Dec '24) scores 8.5 on the Artificial Analysis Intelligence Index, ranking 488th of 744 measured models. The closest models on this index are Devstral 2 (8.6), DeepSeek R1 Distill Qwen 32B (8.4), and GLM-4.6V (8.4). At $0.463 per million tokens (3:1 input/output blended price), it is priced close to GLM-4.6V ($0.45). With an output speed of 37 t/s, it is slower than Devstral 2 (139 t/s) and GLM-4.6V (82.1 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Devstral 2
Mistral · — · 139 t/s
8.6+0.1—139 t/sCompare - DeepSeek R1 Distill Qwen 32B
DeepSeek · — · —
8.4−0.1——Compare - GLM-4.6V
Z.AI · $0.45 · 82.1 t/s
8.4−0.1$0.4582.1 t/sCompare - GPT-4o (2024-11-20)
OpenAI · $4.38 · —
8.4−0.1$4.38—Compare
