DeepSeek V4 Pro 0423
✓ VERIFIED BY DEEPSEEKDeepSeek · Open weights · Released Apr 24, 2026 · deepseek/deepseek-v4-pro
Best $0.661 in / $1.32 out via Baidu · fp8
1.05M contextToolsReasoningStructured outputPrompt cachingOpen weights
Available from 15 providersUSD / MTOK · PRICES 15 MINUTES AGO · SPEED 1 MINUTE AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | SMOKE | UPTIME |
|---|---|---|---|---|---|---|---|---|---|
| $0.661 | $1.32 | $0.055 | 1.05M | fp8 | 0.88s° | 74° tok/s | 3/5 | 100.0% | |
| $0.662 | $1.32 | $0.055 | 1.02M | fp8 | 3.2s° | 64° tok/s | 4/5 | 100.0% | |
| $0.90 | $1.50 | $0.20 | 1.05M | fp4 | 1.2s° | 111° tok/s | — | 100.0% | |
| $0.90 | $1.70 | $0.08 | 1.05M | fp8 | 1.2s° | 48° tok/s | — | 99.4% | |
| $1.04 | $2.08 | $0.08 | 1.05M | fp8 | — | — | — | 100.0% | |
| $1.04 | $2.09 | $0.209 | 1.05M | — | 1.4s° | 33° tok/s | 5/5 | 100.0% | |
| $1.28 | $2.56 | $0.108 | 1.05M | fp8 | 1.7s° | 67° tok/s | — | 100.0% | |
| $1.30 | $2.60 | $0.10 | 1.05M | fp8 | 0.98s° | 57° tok/s | — | 99.9% | |
| $1.31 | $3.56 | $1.31 | — | — | — | — | — | — | |
| $1.32 | $3.96 | $0.044 | 1M | — | 1.2s | 76 tok/s | — | — | |
| $1.42 | $2.83 | $0.118 | 1M | fp8 | 1.5s° | 67° tok/s | — | 99.9% | |
| $1.56 | $3.12 | $0.13 | 1M | fp8 | — | — | — | 100.0% | |
| $1.65 | $3.30 | $0.33 | 1M | — | 1.6s° | 66° tok/s | — | 99.3% | |
| $1.91 | $3.83 | $0.16 | 1.05M | — | 1.9s° | 207° tok/s | — | 100.0% | |
| $1.95 | $3.90 | $0.15 | 1.05M | fp8 | — | — | — | 100.0% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured against the provider API directly; ° = routed via OpenRouter (adds a hop) · ✕ = probed, endpoint returned no measurable stream · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.2 · methodology v1.1 · measured 1 minute ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Quality benchmarksCURATED AUG 24, 2026
Third-party reported results, hand-curated with sources — not measured by ModelIndex. Our own measured serving data (TTFT/TPS/SMOKE) is in the providers table above.
Input price history · best listed price, observed
↓ 30% SINCE SEP 25, 2026
$0.94 · SEP 25, 2026$0.661 · TODAY
Pricing variants · USD / MTok in / out
STANDARD
$0.661 / $1.32
per MTok in / out
CACHE READ $0.055CACHE WRITE Free
Recent changesRSS ↗
Nov 21, 2026endpoint removed · atlascloud17 → 15
Nov 21, 2026endpoint removed · gmicloud17 → 15
Nov 16, 2026price increased · input0.2079 → 0.9
Nov 16, 2026price increased · output0.4157 → 1.5
Nov 14, 2026price decreased · input0.9 → 0.4124
Nov 14, 2026price decreased · output1.5 → 0.8247
Nov 10, 2026endpoint removed · nebius16 → 15
Nov 9, 2026price increased · input0.4124 → 0.9
About
DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...
Specifications
Context window1,048,576
Max output393,216
Modalities intext
Modalities outtext
LicenseOpen weights
ReleasedApr 24, 2026
TokenizerDeepSeek
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/deepseek/deepseek-v4-pro)Sources
deepseek ↗15 minutes ago
OFFICIAL PRICINGOpenRouter API15 minutes ago
AGGREGATORLiteLLM dataset15 minutes ago
AGGREGATORHugging Face15 minutes ago
COMMUNITYModelIndex benchmarks1 minute ago
MEASURED✓ Verified against DeepSeek’s own source.
Verified 15 minutes ago · list prices, not negotiated rates.