DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes
Specification
deepseek listing
Context
1M
Max output
384K
Released
2026-08-12
Knowledge cutoff
2025-05
Retires
2028-02-20
Open weights
yes
Input
text
Output
text
Price
US dollars per million tokens · deepseek list price
Input
$0.435
Output
$0.87
Cache read
$0.0036
Cache write
free
Batch and priority prices
US dollars per million tokens · per listing
| Price | Listing | $/M |
|---|---|---|
| Priority input | fireworks_ai | $1.20 |
| Priority input | fireworks_ai | $1.65 |
| Priority output | fireworks_ai | $1.20 |
| Priority output | fireworks_ai | $4.95 |
Quality
1 benchmark · 1 source
Composite
35th
percentile of 334 scored models
Rank
217 / 334
±0.53 sd
Evidence
1 × 1
too little evidence to place with confidence
Effort range
—
one configuration only
general25th1/1 bench
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| LiveBench | 72.6 | default · 23 runs | LiveBench | 2026-06-25 |
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 91 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| deepseekvendor | $0.435 $1.32 | $0.87 $3.96 | $0.0036 $0.044 | free | 1M | 384K 393K | — |
| alibaba-token-plan | free | free | free | free | 1M | 384K | — |
| alibaba-token-plan-cn | free | free | free | free | 1M | 384K | — |
| kenari | free | free | — | — | 1M | 384K | — |
| nvidia · deepseek-ai | free $0.435 | free $0.87 | $0.0036 | — | 1M 1M | 384K 393K | — |
| scnet-token-plan | free | free | free | — | 1M | 384K | — |
| sensenova | free | free | free | — | 1M | 66K | — |
| unorouter · free | free | free | — | — | 1M | 384K | — |
| volcengine-coding-plan | free | free | free | — | 1M | 384K | — |
| routing-run | $0.348 | $0.696 | — | — | 1M | 64K | — |
| crof | $0.35 | $0.80 | $0.01 $0.003 | — | 1M | 131K | — |
| ebcloud | $0.4286 | $0.8571 | — | — | 1M | 384K | — |
| alibaba-cn | $0.435 | $0.87 | $0.0036 | — | 1M | 384K | — |
| auriko | $0.435 | $0.87 | $0.0036 | — | 1M | 384K | — |
| llmgateway | $0.435 | $0.87 | $0.0036 | — | 1.1M | 384K | — |
| llmgateway-providers · tencent | $0.435 | $0.87 | $0.0036 | — | 1M | 393K | — |
| modelis | $0.435 | $0.87 | — | — | 1M | 384K | — |
| pioneer · deepseek-ai | $0.435 | $0.87 | $0.0036 | $0.435 | 1M | 131K | — |
| tencent · tencent | $0.435 | $0.87 | $0.0036 | free | 1M | 384K | — |
| tokengo | $0.435 | $0.87 | — | — | 1M | 384K | — |
| zenmux | $0.435 | $0.87 | $0.0036 | — | 1M | 384K | — |
| orcarouter | $0.442 | $0.884 | $0.06 | — | 1M | 384K | — |
| runinfra · deepseek-ai | $0.60 | $1.90 | $0.03 | — | 1M | 33K | — |
| edenai | $0.66 $1.32 $1.30 | $1.98 $3.96 $2.60 | $0.066 $1.32 $0.044 $0.132 $0.10 $0.13 | $1.32 | 1M 979K 1M | 384K | — |
| ollama-cloud | $0.66 | $1.98 | $0.022 | — | 1M | 1M | — |
| opencode-go | $0.66 | $1.98 | $0.022 | — | 1M | 384K | — |
| vercel | $0.66 | $1.98 | $0.066 $0.022 | — | 1M | 384K | — |
| aihubmix | $0.6918 | $2.0754 | $0.0231 | — | 1M | 384K | — |
| above | $0.726 | $2.178 | $0.0242 | — | 1M | 384K | — |
| unorouter | $0.8999 | $1.7999 | — | — | 1M | 384K | — |
| llmgateway-providers · runware | $0.961 | $1.922 | $0.079 | — | 1M | 384K | — |
| openrouter | $0.9834 $1.60 $0.8599 $0.5795 | $2.9502 $3.20 $1.7198 $1.7384 | $0.0328 $0.135 $0.0717 $0.0193 | — | 1M | 384K 393K | — |
| aiand · deepseek-ai | $1.00 | $2.50 | $0.25 | — | 1M | 384K | — |
| neuralwatt | $1.00 | $3.00 | $0.10 | — | 1M | 393K | — |
| nano-gpt | $1.10 | $2.50 $2.20 | $0.04 $0.11 | — | 1M | 384K | — |
| nano-gpt · thinking | $1.10 | $2.50 $2.20 | $0.04 $0.11 | — | 1M | 384K | — |
| wandb · wandb | $1.15 $1.31 | $2.55 $3.96 | $0.20 $0.044 | — | 1M | — | — |
| melious | $1.1592 $1.8547 | $3.4776 $3.7094 | $0.1159 $0.4637 | — | 1M | 384K | — |
| fireworks_ai | $1.20 $1.32 | $1.20 $3.96 | $0.60 $0.044 | — | 1M | 384K 131K | — |
| crossmodel | $1.215 | $3.645 | $0.0405 | $1.215 | 1M | 384K | — |
| deepinfra | $1.30 | $2.60 | $0.10 | — | 1M | — | — |
| deepinfra · deepseek-ai | $1.30 | $2.60 | $0.10 | — | 1M | 384K 16K | — |
| wandb · deepseek-ai | $1.31 $1.15 | $3.96 $2.55 | $0.044 $0.20 | — | 1M | 1M | — |
| arcee | $1.32 $1.74 | $3.96 $3.48 | $0.044 $0.20 | — | 1M 512K | 384K | — |
| baseten · deepseek-ai | $1.32 $1.74 | $3.96 $3.48 | $0.145 | — | 1M | 262K | — |
| cloudflare-workers-ai · cf | $1.32 | $3.96 | $0.044 | — | 1M | 1M | — |
| databricks | $1.32 | $3.96 | $0.132 | $1.32 | 1M | 393K | — |
| digitalocean | $1.32 $0.87 | $3.96 $1.74 | $0.044 $0.174 | — | 1M | 1M 384K | — |
| empiriolabs | $1.32 $1.65 | $3.96 $3.30 | $1.32 $1.65 | — | 1M | 393K | — |
| fireworks-ai · accounts | $1.32 | $3.96 | $0.044 | — | 1M | 384K | — |
| huggingface · deepseek-ai | $1.32 $0.435 | $3.96 $0.87 | $0.0036 | — | 1M 1M | 384K 393K | — |
| kilo | $1.32 $1.60 | $3.96 $3.20 | $0.044 $0.135 | — | 1M 1M | 384K 393K | — |
| llmgateway-providers · baidu | $1.32 | $3.96 | $0.132 | — | 1M | 131K | — |
| llmgateway-providers · bytedance | $1.32 | $3.96 | $0.044 | — | 1M | 393K | — |
| novita | $1.32 $1.60 | $3.96 $3.20 | $0.132 $0.135 | — | 1M | 393K | — |
| ofox | $1.32 | $3.96 | $0.044 $0.15 | — | 1M | 384K | — |
| requesty | $1.32 | $3.96 | $0.044 | — | 1M | 131K | — |
| togetherai · deepseek-ai | $1.32 $1.74 | $3.96 $3.48 | $0.13 $0.20 | — | 1M 512K | 384K | — |
| vivgrid | $1.35 $0.435 | $3.00 $0.87 | $0.05 $0.0036 | — | 1M | 384K | — |
| gmicloud · deepseek-ai | $1.392 | $2.784 | $0.116 | — | 1M | 384K | — |
| hyper | $1.4372 $2.40 | $4.3116 $4.80 | $0.0479 $0.20 | — | 1M | 262K 384K | — |
| jalapeno | $1.60 | $3.38 | — | — | 1M | 384K | — |
| novita-ai | $1.60 | $3.20 | $0.135 | — | 1M | 393K | — |
| bothub | $1.61 | $4.84 | — | — | 1M | 384K | — |
| venice | $1.65 | $4.95 $3.301 | $0.165 $0.33 | — | 1M | 33K | — |
| abacus · deepseek-ai | $1.74 | $3.48 | $0.15 | — | 1M | 33K | — |
| azure | $1.74 | $3.48 | — | — | 1M | 384K | — |
| azure_ai | $1.74 | $3.48 | $0.145 | — | 1M | 384K | 2028-02-20 |
| cline-pass · cline-pass | $1.74 | $3.48 | $0.0145 | — | 1M | 384K | — |
| cloudflare-ai-gateway | $1.74 | $3.48 | $0.145 | — | 131K | 384K | — |
| fastrouter | $1.74 | $3.48 | — | — | 1M | 384K | — |
| frogbot | $1.74 | $3.48 | $0.14 | — | 128K | 8K | — |
| hpc-ai | $1.74 | $3.48 | $0.145 | — | 1M | 128K | — |
| impossibl | $1.74 | $3.48 | $0.145 | — | 1M | 384K | — |
| llmgateway-providers · canopywave | $1.74 | $3.48 | $0.01 | — | 1M | 393K | — |
| merge-gateway | $1.74 $1.65 | $3.48 $3.30 | $0.0036 $0.145 | — | 1M | 384K 393K 262K | — |
| opencode | $1.74 | $3.84 | $0.145 | — | 1M | 384K | — |
| siliconflow · deepseek-ai | $1.74 | $3.48 | $0.145 | — | 1M | 384K | — |
| siliconflow-cn · deepseek-ai | $1.74 | $3.48 | $0.145 | — | 1M | 393K | — |
| together_ai | $1.74 $1.32 | $3.48 $3.96 | $0.20 $0.13 | — | 512K 1M | — | 2026-08-27 |
| nebius | $1.75 | $3.50 | — | — | 1M | 1M | — |
| nebius · deepseek-ai | $1.75 | $3.50 | $0.15 | — | 1M | 1M | — |
| tensorx | $1.75 | $3.50 | $0.4375 | $2.185 | 1M | 384K | — |
| cortecs | $2.00 $1.73 | $3.999 $3.46 | $0.50 $0.432 | — | 1M | 64K 1M | — |
| edenai · tensorx | $2.00 | $4.00 | $0.50 | — | 1M | 384K | — |
| dashscope · dashscope | $2.40 | $4.80 | $0.20 | — | 1M | 393K | — |
| llmgateway-providers | $2.40 $1.32 $1.30 $0.435 | $4.80 $3.96 $2.60 $0.87 | $0.20 $0.13 $0.10 $0.0036 $0.044 | — | 1M 1M 1.1M | 393K 164K 64K | — |
| qwen_ai_platform · qwen-ai-platform | $2.40 | $4.80 | $0.20 | — | 1M | 393K | — |
| qwencloud · qwencloud | $2.40 | $4.80 | $0.20 | — | 1M | 393K | — |
| anyapi | — | — | — | — | 1M | 384K | — |
| model-oracle-ai | — | — | — | — | 1M | 384K | — |
Price history
append-only observations · a listing writes a row only when its price moves
| Date | Host | Field | $/M |
|---|---|---|---|
| 2026-09-02 | edenai | Cache read | $0.044 |
| 2026-09-04 | edenai | Cache read | $0.044 |
| 2026-09-11 | edenai | Cache read | $1.32 |
| 2026-09-15 | edenai | Cache read | $0.066 |
| 2026-09-02 | fireworks_ai | Cache read | $0.145 |
| 2026-09-15 | fireworks_ai | Cache read | $0.60 |
| 2026-09-02 | kilo | Cache read | $0.132 |
| 2026-09-09 | kilo | Cache read | $0.044 |
| 2026-09-02 | llmgateway-providers | Cache read | $0.044 |
| 2026-09-04 | llmgateway-providers | Cache read | $0.20 |
| 2026-09-02 | merge-gateway | Cache read | $0.022 |
| 2026-09-15 | merge-gateway | Cache read | $0.0036 |
| 2026-09-02 | nano-gpt · thinking | Cache read | $0.11 |
| 2026-09-04 | nano-gpt · thinking | Cache read | $0.04 |
| 2026-09-02 | ofox | Cache read | $0.044 |
| 2026-09-12 | ofox | Cache read | $0.044 |
| 2026-09-02 | openrouter | Cache read | $0.022 |
| 2026-09-02 | openrouter | Cache read | $0.0372 |
| 2026-09-02 | openrouter | Cache read | $0.022 |
| 2026-09-03 | openrouter | Cache read | $0.0372 |
| 2026-09-05 | openrouter | Cache read | $0.0374 |
| 2026-09-06 | openrouter | Cache read | $0.0374 |
| 2026-09-07 | openrouter | Cache read | $0.0374 |
| 2026-09-08 | openrouter | Cache read | $0.035 |
| 2026-09-09 | openrouter | Cache read | $0.0579 |
| 2026-09-10 | openrouter | Cache read | $0.035 |
| 2026-09-12 | openrouter | Cache read | $0.0184 |
| 2026-09-13 | openrouter | Cache read | $0.0184 |
| 2026-09-14 | openrouter | Cache read | $0.0352 |
| 2026-09-15 | openrouter | Cache read | $0.0328 |
| 2026-09-02 | togetherai · deepseek-ai | Cache read | $0.20 |
| 2026-09-04 | togetherai · deepseek-ai | Cache read | $0.13 |
| 2026-09-02 | vivgrid | Cache read | $0.0036 |
| 2026-09-08 | vivgrid | Cache read | $0.05 |
| 2026-09-02 | wandb · deepseek-ai | Cache read | $0.20 |
| 2026-09-04 | wandb · deepseek-ai | Cache read | $0.044 |
| 2026-09-02 | wandb · wandb | Cache read | $0.20 |
| 2026-09-11 | wandb · wandb | Cache read | $0.20 |
| 2026-09-02 | baseten · deepseek-ai | Input | $1.74 |
| 2026-09-04 | baseten · deepseek-ai | Input | $1.32 |
| 2026-09-02 | edenai | Input | $1.32 |
| 2026-09-04 | edenai | Input | $0.5808 |
| 2026-09-15 | edenai | Input | $0.66 |
| 2026-09-02 | fireworks_ai | Input | $1.74 |
| 2026-09-15 | fireworks_ai | Input | $1.20 |
| 2026-09-02 | huggingface · deepseek-ai | Input | $0.435 |
| 2026-09-04 | huggingface · deepseek-ai | Input | $1.32 |
| 2026-09-02 | llmgateway-providers | Input | $1.32 |
| 2026-09-04 | llmgateway-providers | Input | $2.40 |
| 2026-09-02 | merge-gateway | Input | $0.66 |
| 2026-09-15 | merge-gateway | Input | $1.74 |
| 2026-09-02 | openrouter | Input | $0.66 |
| 2026-09-02 | openrouter | Input | $1.1154 |
| 2026-09-02 | openrouter | Input | $0.66 |
| 2026-09-03 | openrouter | Input | $1.1154 |
| 2026-09-05 | openrouter | Input | $1.1207 |
| 2026-09-06 | openrouter | Input | $1.1207 |
| 2026-09-07 | openrouter | Input | $1.1207 |
| 2026-09-08 | openrouter | Input | $1.0494 |
| 2026-09-09 | openrouter | Input | $0.5795 |
| 2026-09-10 | openrouter | Input | $1.0494 |
| 2026-09-12 | openrouter | Input | $0.5782 |
| 2026-09-13 | openrouter | Input | $0.5782 |
| 2026-09-14 | openrouter | Input | $1.0547 |
| 2026-09-15 | openrouter | Input | $0.9834 |
| 2026-09-02 | togetherai · deepseek-ai | Input | $1.74 |
| 2026-09-04 | togetherai · deepseek-ai | Input | $1.32 |
| 2026-09-02 | vivgrid | Input | $0.435 |
| 2026-09-08 | vivgrid | Input | $1.35 |
| 2026-09-02 | wandb · deepseek-ai | Input | $1.15 |
| 2026-09-04 | wandb · deepseek-ai | Input | $1.31 |
| 2026-09-02 | wandb · wandb | Input | $1.15 |
| 2026-09-11 | wandb · wandb | Input | $1.15 |
| 2026-09-02 | baseten · deepseek-ai | Output | $3.48 |
| 2026-09-04 | baseten · deepseek-ai | Output | $3.96 |
| 2026-09-02 | edenai | Output | $3.96 |
| 2026-09-04 | edenai | Output | $1.7424 |
| 2026-09-15 | edenai | Output | $1.98 |
| 2026-09-02 | fireworks_ai | Output | $3.48 |
| 2026-09-15 | fireworks_ai | Output | $1.20 |
| 2026-09-02 | huggingface · deepseek-ai | Output | $0.87 |
| 2026-09-04 | huggingface · deepseek-ai | Output | $3.96 |
| 2026-09-02 | llmgateway-providers | Output | $3.96 |
| 2026-09-04 | llmgateway-providers | Output | $4.80 |
| 2026-09-02 | merge-gateway | Output | $1.98 |
| 2026-09-15 | merge-gateway | Output | $3.48 |
| 2026-09-02 | nano-gpt · thinking | Output | $2.20 |
| 2026-09-04 | nano-gpt · thinking | Output | $2.50 |
| 2026-09-02 | openrouter | Output | $1.98 |
| 2026-09-02 | openrouter | Output | $3.3462 |
| 2026-09-02 | openrouter | Output | $1.98 |
| 2026-09-03 | openrouter | Output | $3.3462 |
| 2026-09-05 | openrouter | Output | $3.362 |
| 2026-09-06 | openrouter | Output | $3.362 |
| 2026-09-07 | openrouter | Output | $3.362 |
| 2026-09-08 | openrouter | Output | $3.1482 |
| 2026-09-09 | openrouter | Output | $1.7384 |
| 2026-09-10 | openrouter | Output | $3.1482 |
| 2026-09-12 | openrouter | Output | $1.7345 |
| 2026-09-13 | openrouter | Output | $1.7345 |
| 2026-09-14 | openrouter | Output | $3.164 |
| 2026-09-15 | openrouter | Output | $2.9502 |
| 2026-09-02 | togetherai · deepseek-ai | Output | $3.48 |
| 2026-09-04 | togetherai · deepseek-ai | Output | $3.96 |
| 2026-09-02 | vivgrid | Output | $0.87 |
| 2026-09-08 | vivgrid | Output | $3.00 |
| 2026-09-02 | wandb · deepseek-ai | Output | $2.55 |
| 2026-09-04 | wandb · deepseek-ai | Output | $3.96 |
| 2026-09-02 | wandb · wandb | Output | $2.55 |
| 2026-09-11 | wandb · wandb | Output | $2.55 |
Other listings of this model
same model, different key
More from DeepSeek
most-hosted first