D
DeepSeek V4 Pro
The 1.6-trillion-parameter former flagship — the vendor is already migrating users to V4.1 Flash
Intelligence v5
23.0
Leaderboard rank
#17
Input / output
$1.32/$3.96
Blended 3:1
$1.98
Output speed
72 tok/s
Value
12 pts/$
Capability profile Normalised to the field maximum
Specifications
Vendor
DeepSeek
Released
2026-08-13
Weights
Open (MIT)
Context
1,048,576 tokens
Max output
Undisclosed
Parameters
1600B total / 49B active
Input price
$1.32 / 1M
Output price
$3.96 / 1M
Cache hit
$0.04 / 1M
Output speed
72 tok/s
Intelligence
23.0（v5）
Benchmark scores
How it is measured →Ranked 17 · bars are scaled to the field maximum for that benchmark.
| Benchmarks | Score | Vs. best in field | Gap |
|---|---|---|---|
| AA Intelligence Index v5 Intelligence index | 23.0 | 39% | −36.4 pts |
| AA Intelligence Index v4.3.2 | — | — | not indexed |
| GPQA Diamond | — | — | not indexed |
| Terminal-Bench 4.0 | — | — | not indexed |
| LiveBench Design | — | — | not indexed |
| Vals Index | — | — | not indexed |
Assessment
1.6T total parameters with 49B active. DeepSeek cites reproducibly-tested third-party results showing V4.1 Flash has overtaken V4 Pro, and says it is retiring the latter. Still MIT licensed; off-peak pricing is $0.66/$1.98.
At the same price
The four models closest on blended price, side by side.
| Model | v5 | Blended | tok/s | Score gap |
|---|---|---|---|---|
| GPT-6.1 TerraOpenAI | 51.0 | $2.00 | 94 | +28.0 |
| GLM-5.3 Open weightsZ.ai | 32.0 | $1.00 | 77 | +9.0 |
| Grok 4.7xAI | 35.0 | $3.00 | 105 | +12.0 |
| MiMo-V2.6 Pro Open weightsXiaomi | 30.0 | $0.61 | 81 | +7.0 |
Compare side by side
Line it up against other models on price, speed and every benchmark.
The comparison tool lines up to 4 models side by side with switchable normalisation — built for purchasing decisions.
Open the comparison tool → Back to leaderboard