DeepSeek R1
Azure AI Foundry chat Deprecated 83d agomodel fallback routing to swap models without code changes.
DeepSeek R1 is an Azure AI Foundry chat model.Input is priced at $1.35/M tokens and output at $5.40/M tokens. Route DeepSeek R1 via Future AGI's Agent Command Center for unified observability, caching, and 15 routing strategies including cost-optimized fallback.
Estimate DeepSeek R1 spend
Pick a workload, fine-tune the sliders, and see the monthly bill.
Estimate uses $1.35/M input · $5.40/M output. Provider pricing changes. Production costs vary with retries, streaming overhead, and tool-call rounds.
Want this for free? Cache + route via Agent Command Center — first 100K requests and 100K cache hits free every month.
Pricing
Per-token rates, expressed in USD per 1M tokens. Verified Oct 22, 2026.
| Input | $1.35/M | |
| Output | $5.40/M | |
| Blended (3:1 in:out) | $2.36/M Cross-model comparable price |
Context & limits
- Context window
- —
- Max input
- —
- Max output
- —
- Modalities
- text
Capabilities
- Function calling — not advertised
- Parallel tool calls — not advertised
- Vision input — not advertised
- Audio input — not advertised
- Audio output — not advertised
- PDF input — not advertised
- Streaming ✓ supported
- Structured output — not advertised
- Prompt caching — not advertised
- Reasoning — not advertised
Strengths & caveats — when to pick DeepSeek R1
Data-driven from DeepSeek R1's percentile within chat peers.
Where it's strong
Watch out for
- !agentic workflows — no advertised function-calling; use a tool-capable model and route via Agent Command Center for fallback
- !strict structured output — no JSON-schema enforcement, expect retry loops
- !already deprecated — provider stopped accepting new traffic 83 days ago
Benchmark scores
Reported public benchmark numbers. Each row links to the source. Faded bar shows 6-peer average for context.
Call DeepSeek R1 via Agent Command Center
One OpenAI-compatible endpoint. Routing, fallback, semantic caching, guardrails, and cost tracking come along for the ride. First 100K requests + 100K cache hits free every month.
agentcc / @agentcc/client). Per-call metadata — provider, cost, latency, cache hit, request id — is returned on x-agentcc-* response headers, so any HTTP client can read it.# DeepSeek R1 via the Agent Command Center Python SDK
# pip install agentcc
import os
from agentcc import AgentCC
client = AgentCC(
api_key=os.environ["AGENTCC_API_KEY"], # from app.futureagi.com → Settings → API Keys
base_url="https://gateway.futureagi.com/v1",
)
resp = client.chat.completions.create(
model="azure-ai-foundry/deepseek-r1",
messages=[{"role": "user", "content": "Hello, DeepSeek R1!"}],
)
print(resp.choices[0].message.content)
print(f"Tokens: {resp.usage.total_tokens}")
# Per-call gateway metadata is returned on x-agentcc-* response headers.
# When you need it programmatically, use .with_raw_response to get them:
raw = client.chat.completions.with_raw_response.create(
model="azure-ai-foundry/deepseek-r1",
messages=[{"role": "user", "content": "Same call, but I want the headers."}],
)
print("Provider:", raw.headers.get("x-agentcc-provider"))
print("Latency:", raw.headers.get("x-agentcc-latency-ms"), "ms")
print("Cost: ", raw.headers.get("x-agentcc-cost"), "USD")
print("Cache: ", raw.headers.get("x-agentcc-cache"))AGENTCC_API_KEY with a key fromapp.futureagi.com.Gateway docs ↗Advanced: fallback + cache config (YAML) ▸
strategy: cost-optimized
targets:
- model: deepseek-r1
provider: azure-ai-foundry
weight: 80
fallbacks:
- model: deepseek-v4-pro
provider: dashscope
- model: claude-sonnet-4-5
provider: anthropic
guardrails: [pii, prompt-injection, secrets]
cache: { exact: true, semantic: true }Same model on other providers
deepseek-r1 is also available via 3 other routes. Pricing, regions, and capabilities can differ — compare before routing production traffic.
| Provider | Input / 1M | Output / 1M | Verified |
|---|---|---|---|
| Snowflake Cortex | $1.35/M | $5.40/M | Oct 22, 2026 |
| DeepSeek | $0.550/M | $2.19/M | Oct 22, 2026 |
| SambaNova | $5.00/M | $7.00/M | Oct 22, 2026 |
Explore similar models
Grouped by Chatbot Arena tier (DeepSeek R1 sits at 1398 ELO). Side-by-side links appear where a comparison page exists; otherwise, open the model profile.
- GPT-5.4Azure AI Foundry · $2.50/M in · $15.00/M out · 922,000 ctxCompare side-by-side →
- Claude Haiku 4.5Azure AI Foundry · $1.00/M in · $5.00/M out · 200,000 ctxCompare side-by-side →
- DeepSeek v3.2Azure AI Foundry · $0.580/M in · $1.68/M out · 128,000 ctxCompare side-by-side →
- Mistral Large latestAzure AI Foundry · $2.00/M in · $6.00/M out · 128,000 ctxCompare side-by-side →
- DeepSeek V4 ProAlibaba DashScope · $2.40/M in · $4.80/M out · 1,000,000 ctxCompare side-by-side →
- Gemini 3.5 Flash LiteGoogle Vertex AI · $0.300/M in · $2.50/M out · 1,048,576 ctxCompare side-by-side →
- Grok 4.6xAI · $2.00/M in · $6.00/M out · 500,000 ctxCompare side-by-side →
- Grok 4.7xAI · $2.00/M in · $6.00/M out · 500,000 ctxCompare side-by-side →
≥30 ELO higher
- Gemini 4 ArgonGoogle AI · $2.00/M in · $10.00/M out · 1,048,576 ctxCompare side-by-side →
- GPT-5.5 (2026-04-23)OpenAI · $5.00/M in · $30.00/M out · 1,050,000 ctxCompare side-by-side →
- GPT-5.5OpenAI · $5.00/M in · $30.00/M out · 1,050,000 ctxCompare side-by-side →
- GPT-5.4Azure AI Foundry · $2.50/M in · $15.00/M out · 922,000 ctxCompare side-by-side →
25–100 ELO lower, ≤50% of price
- Mistral Large latestMistral AI · $0.500/M in · $1.50/M out · 262,144 ctxCompare side-by-side →
- Gemini 1.5 FlashGoogle Vertex AI · $0.0750/M in · $0.300/M out · 1,000,000 ctxView model →
- GPT-4.1 nano (2025-04-14)Azure OpenAI · $0.1000/M in · $0.400/M out · 1,047,576 ctxView model →
- GPT-4o miniAzure OpenAI · $0.150/M in · $0.600/M out · 128,000 ctxView model →
FAQ
How much does DeepSeek R1 cost?
Input is priced at $1.35 per 1M tokens and output at $5.40 per 1M tokens (Azure AI Foundry, last verified Oct 22, 2026).
What is the context window of DeepSeek R1?
Context window for DeepSeek R1 is not currently public.
Does DeepSeek R1 support function calling?
DeepSeek R1 does not currently advertise function-calling support. For agentic workloads, prefer a tool-calling-capable model and route via Agent Command Center for fallback.
Is DeepSeek R1 good for production?
DeepSeek R1 is best evaluated against your own production traces. Pipe traffic through Agent Command Center to compare it head-to-head against alternatives in shadow mode.
How can I route to DeepSeek R1 with fallback?
Use Agent Command Center: a single OpenAI-compatible endpoint that supports cost-optimized routing, latency-aware retries, model fallback, and shadow traffic. Configure once, swap models without app changes.
Useful links for DeepSeek R1
Official sources, independent benchmarks, and pricing aggregators — no random search-engine guesses.
Third-party evals — verify the marketing.
Cross-check our number against the rest of the ecosystem.