LLM Pricing
PricingLeaderboardsToolsProvidersReleasesGuides

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

DeepSeek V3.1 Terminus

deepseek·deepseek/deepseek-v3.1-terminus·GA·Closed·deepseek series
DeepSeek chat model for instruction following, coding, and analysis

DeepSeek V3.1 Terminus is offered by 12 providers on this page. Public prices are shown for 11 of them. Its reference price is $0.27 per 1M input tokens and $1.00 per 1M output tokens.

Artificial Analysis rates it 14.8 on the Intelligence Index, with 43.5 for coding and 6.8 for agentic tasks. Against its lowest blended price of $0.362 per 1M, that is roughly 41 index points per dollar, which is the value ratio the leaderboards rank on.

The context window is 163,840 tokens at the reference host, but hosts report different limits, from 128,000 to 164,000, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use. Its training knowledge cuts off at 2025-03-31.

Specs & pricing

Input / output per 1M tokens
Reference price·Alibaba (China)
$0.27 / $1.00
Blended $0.45 · Cache read —
Lowest paid·NanoGPTGateway
$0.25 / $0.70
Blended $0.36 · 1.1× spread
Context
163,840
Output limit
65,536
Knowledge cutoff
2025-03-31
Released / updated
2025-09-22 / 2025-11-25
Capabilities
✓ Reasoning✓ Tool use✓ Structured output✓ TemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
Text

Quality & performanceArtificial Analysis · Intelligence Index v4.3 · rep. tier Reasoning

Intelligence14.8
Value41
Coding43.5
Agentic6.8
Output speed—
TTFT—
Value formulaIQ 14.8 ÷ min blended $0.362 = 41
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
Non-reasoningIQ 13.9
ReasoningIQ 14.8

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 12 providers11 with public prices

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
NanoGPT
deepseek-ai/DeepSeek-V3.1-Terminus
Gateway$0.25$0.70$0.13—128,000 ⚠65,536
Abacus
deepseek-ai/DeepSeek-V3.1-Terminus
Gateway$0.27$1.00——128,000 ⚠8,192
NovitaAIGateway$0.27$1.00$0.14—131,072 ⚠32,768
OpenRouterGateway$0.27$1.00——163,840147,456
Vercel AI GatewayCloud$0.27$1.00$0.14—131,072 ⚠32,768
HeliconeGateway$0.27$1.00$0.22—128,000 ⚠16,384
SiliconFlow
deepseek-ai/DeepSeek-V3.1-Terminus
Cloud$0.27$1.00——164,000 ⚠164,000
SiliconFlow (China)
deepseek-ai/DeepSeek-V3.1-Terminus
Cloud$0.27$1.00——164,000 ⚠164,000
Kilo GatewayGateway$0.27$1.00——163,840147,456
SiliconFlow (China)
Pro/deepseek-ai/DeepSeek-V3.1-Terminus
Cloud$0.27$1.00——164,000 ⚠164,000
Alibaba (China)First-party$0.27$1.00——163,84065,536
QiniuGateway————128,000 ⚠32,000hostTable.undisclosed

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

Toggle (on / off)effort = noneeffort = loweffort = mediumeffort = highbudget_tokens ≥ 128 ≤ 32,768

2 / 12 providers expose no reasoning control (reasoning_options: []).

Your usage cost

1NanoGPT$70.00
2NovitaAI$87.80
3Vercel AI Gateway$87.80
4Helicone$97.52
5Abacus$104.00
6OpenRouter$104.00
The cheapest paid channel is the only channel.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$1.00$02026-08-052026-10-032026-08-05 · Input list $0.272026-08-06 · Input list $0.272026-08-07 · Input list $0.272026-08-08 · Input list $0.272026-08-09 · Input list $0.272026-08-10 · Input list $0.272026-08-11 · Input list $0.272026-08-12 · Input list $0.272026-08-13 · Input list $0.272026-08-18 · Input list $0.272026-09-03 · Input list $0.272026-09-29 · Input list $0.302026-10-03 · Input list $0.272026-08-05 · Output list $1.002026-08-06 · Output list $1.002026-08-07 · Output list $1.002026-08-08 · Output list $1.002026-08-09 · Output list $1.002026-08-10 · Output list $1.002026-08-11 · Output list $1.002026-08-12 · Output list $1.002026-08-13 · Output list $1.002026-08-18 · Output list $1.002026-09-03 · Output list $1.002026-09-29 · Output list $1.002026-10-03 · Output list $1.002026-08-05 · Min blended $0.362026-08-06 · Min blended $0.362026-08-07 · Min blended $0.362026-08-08 · Min blended $0.362026-08-09 · Min blended $0.362026-08-10 · Min blended $0.362026-08-11 · Min blended $0.362026-08-12 · Min blended $0.362026-08-13 · Min blended $0.452026-08-18 · Min blended $0.452026-09-03 · Min blended $0.362026-09-29 · Min blended $0.362026-10-03 · Min blended $0.36

Artificial Analysis evaluations12 items

GPQA Diamond79.2%
Humanity's Last Exam16.4%
MMLU-Pro85.1%
SciCode38.0%
LiveCodeBench79.8%
Terminal-Bench Hard30.3%
Terminal-Bench 2.144.9%
AIME 202589.7%
τ²-Bench Telecom37.1%
τ³-Bench Banking21.0%
AA-LCR69.3%
IFBench57.0%

Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and harness, so rows are not comparable with one another. The Intelligence Index above draws on a different, newer set of evaluations.

Related models

DeepSeek: DeepSeek Pro Latestsame series$0.19 / $4.20DeepSeek V4 Flash Latestsame series$0.015 / $1.28DeepSeek Latestsame series$1.10 / $2.50GLM-5.3-Flashcheaper alternative$0.15 / $0.50DeepSeek V4.1 Flashcheaper alternative$0.15 / $0.60GPT-5 Nanocheaper alternative$0.05 / $0.40Qwen3.8 Flashcheaper alternative$0.15 / $0.47
Data partly from models.dev (MIT) · Alibaba (China) official docs ↗