DeepSeek V3
DeepSeek V3 is worth evaluating for coding, agents, and classification when its provider route and context window match the workload.
Use it for
- Teams evaluating coding, agents, and classification
- Workloads that can use a 64k context window
- Buyers comparing 4 tracked provider routes
Do not use it for
- Vision or document-understanding workloads
- Family
- DeepSeek V3
- Released
- 2024-12-26
- Context
- 64k
- Parameters
- 671B
- Architecture
- Mixture of Experts
- Knowledge cutoff
- 2024-04
- Specialization
- general
- Openness
- Open source
- License
- MITOSI-approvedCommercial use: permitted
- Weights
- Available
- Code
- Unknown
- Training
- Fine-tuned
Cheapest of 13 routes · DeepSeek Platform
About
DeepSeek V3: Latest flagship model. 685B total with MoE. 128K context. Open-source.
DeepSeek V3 is an open-source model. The structured metadata tracks a 64k-token context window, function calling, tool use, and structured outputs. This page tracks provider routes through DeepInfra, Fireworks AI, DeepSeek Platform, and 10 more, with the cheapest tracked route listed at $0.1 input and $0.3 output per 1M tokens. Headline tracked benchmarks include HellaSwag 95.7, HumanEval 85.5, and Massive Multitask Language Understanding 88.5.
Top use-case fit: coding, agents, and build tasks
Coding
Q/$ B4 relevant benchmarks in the decision map.
Agents
Included by capability and metadata signals in the decision map.
Classification
Q/$ B3 relevant benchmarks in the decision map.
Provider price ladder
Compare all 13Compare API pricing across 4 providers for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Route |
|---|---|---|---|
| DeepSeek Platform | $0.140 | $0.280 | Serverless |
| Bitdeer AI | $0.100 | $0.300 | Serverless |
| OpenRouter | $0.252 | $0.378 | Serverless |
| SiliconFlow | $0.150 | $0.500 | Serverless |
Available via routers & gateways(16)
LiteLLM
GatewayOpen-source Python SDK and proxy server that unifies 100+ LLM APIs behind a single OpenAI-compatible interface, with load balancing, cost tracking, and configurable failover.
OpenRouter
HybridUnified hybrid gateway to 400+ models from 60+ providers via a single OpenAI-compatible API, with optional auto-routing that selects the best model per prompt.
Portkey
GatewayProduction AI gateway routing to 1,600+ LLMs with failover, load balancing, semantic caching, and guardrails; Apache 2.0 core is fully self-hostable with the complete feature set.
AIRouter
RouterCommercial LLM router that analyzes incoming requests and routes to the optimal model for cost/quality/latency via a drop-in OpenAI-compatible API, with a privacy-preserving embedding mode that avoids sending prompt content.
Amazon Bedrock Intelligent Prompt Routing
RouterAWS Bedrock's native intelligent prompt router that routes prompts between Anthropic Claude model tiers (Haiku/Sonnet) based on predicted task complexity, with no extra per-routing charge.
Azure AI Foundry Model Router
RouterMicrosoft Azure AI Foundry's native model router that uses a trained ML model to route each prompt in real time to the optimal Azure-hosted model, with Balanced/Cost/Quality mode selection and automatic failover.
Capabilities
Benchmark peer barsfor Coding
Benchmark scores(9)
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| HellaSwag | 95.7 | 10-shotObserved 2026-03-06 | — | Source |
| HumanEval | 85.5 | pass@1Observed 2026-03-06 | — | Source |
| Massive Multitask Language Understanding | 88.5 | 5-shotObserved 2026-03-06 | — | Source |
| LiveCodeBench | 49.6 | 2026-04Observed 2026-04-14 | — | Source |
| Aider Polyglot | 48.4 | 2026-04Observed 2026-04-14 | — | Source |
| BigCodeBench | 50.0 | 2025-01 (Instruct Pass@1)Observed 2026-04-14 | — | Source |
| Chatbot Arena | 1302.0 | —Observed 2026-04-15 | — | Source |
| MMLU PRO | 75.9 | —Observed 2026-04-14 | — | Source |
| Mostly Basic Programming Problems+ | 76.0 | —Observed 2026-05-28 | — | Source |
Migration checks
No linked migration route is available for this model yet.
Rankings & picks(1)
Compare DeepSeek V3 with other models
Comparison and alternatives
Browse all comparisons →Show all 79 popular comparisonssorted by 7-day search impressions
Frequently asked questions
What is the context window of DeepSeek V3?
DeepSeek V3 has a context window of 64k tokens.
How much does DeepSeek V3 cost?
DeepSeek V3 pricing ranges from $0.1/1M to $1.45/1M input tokens depending on the provider.
When was DeepSeek V3 released?
DeepSeek V3 was released on 2024-12-26.
Which providers offer DeepSeek V3?
DeepSeek V3 is available from 13 providers: DeepInfra, Fireworks AI, DeepSeek Platform, Microsoft Foundry, OpenRouter, NVIDIA NIM, AWS Bedrock, Together AI, Bitdeer AI, SiliconFlow, Replicate API, GCP Vertex AI, Vercel AI Gateway.
What benchmarks has DeepSeek V3 been tested on?
DeepSeek V3 has been evaluated on 9 benchmarks, including HellaSwag, HumanEval, Massive Multitask Language Understanding, LiveCodeBench, Aider Polyglot.
Cheapest of 13 routes · DeepSeek Platform