o3
o3 is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.
Use it for
- Teams evaluating coding, rag, and agents
- Workloads that can use a 200k context window
- Buyers comparing 3 tracked provider routes
Do not use it for
- Workloads where another current model has stronger sourced task evidence
- Family
- o3
- Released
- 2025-04-16
- Context
- 200k
- Max output
- 100,000
- Architecture
- Decoder Only
- Knowledge cutoff
- 2024-06
- Specialization
- general
- Openness
- Proprietary
- License
- ProprietaryCommercial use: conditional
- Weights
- Not released
- Code
- Unknown
- Training
- Fine-tuned
Cheapest of 3 routes · OpenAI API · cache read $0.500
About
OpenAI o3 reasoning model with advanced multi-step problem-solving capabilities.
o3 is a proprietary model. The structured metadata tracks a 200k-token context window, multimodal input, reasoning, function calling, tool use, structured outputs, and code execution. This page tracks provider routes through OpenAI API, OpenRouter, and Vercel AI Gateway, with the cheapest tracked route listed at $2 input and $8 output per 1M tokens. Headline tracked benchmarks include HumanEval 96.7, SWE-bench Verified 71.7, and LiveCodeBench 85.5.
Top use-case fit: coding, agents, and build tasks
Coding
Q/$ D4 relevant benchmarks in the decision map.
RAG
Included by capability and metadata signals in the decision map.
Agents
Q/$ C1 relevant benchmark in the decision map.
Provider price ladder
Compare all 3Compare API pricing across 3 providers for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Batch in / out | Cache | Route |
|---|---|---|---|---|---|
| OpenAI API | $2.00 | $8.00 | $1.00 / $4.00 | read $0.500 | Serverless |
| OpenRouter | $2.00 | $8.00 | - | - | Serverless |
| Vercel AI Gateway | $2.00 | $8.00 | - | read $0.500 | Serverless |
Available via routers & gateways(15)
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.
Helicone
GatewayObservability-first AI gateway with routing, caching, rate limiting, and request tracing; Apache 2.0 open-source core with a managed hosted tier for logging and analytics.
Kong AI Gateway
GatewayMulti-LLM AI gateway built on Kong Gateway 3.x, adding semantic routing, load balancing, guardrails, and MCP traffic analytics as plugins over Kong's existing API management platform.
Capabilities
Benchmark peer barsfor Coding
Benchmark scores(12)
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| HumanEval | 96.7 | 2025-04Observed 2026-04-09 | — | Source |
| SWE-bench Verified | 71.7 | custom agent scaffold, high reasoningObserved 2025-04-16 | — | Source |
| LiveCodeBench | 85.5 | v6 pass@1, independent Kaggle leaderboardObserved 2025-11-13 | — | Source |
| Aider Polyglot | 81.3 | 2026-04 (high)Observed 2026-04-14 | — | Source |
| Chatbot Arena | 1412.0 | —Observed 2026-04-15 | — | Source |
| Google-Proof Q&A | 87.7 | GPQA Diamond, high reasoningObserved 2025-04-16 | — | Source |
| Massive Multi-discipline Multimodal Understanding | 82.9 | standard MMMUObserved 2025-04-16 | — | Source |
| MathVista | 86.8 | —Observed 2026-06-03 | — | Source |
| MMMU Pro | 76.4 | LLM-Stats aggregator (official no-tools or equivalent)Observed 2026-06-07 | — | Source |
| AIME 2024 | 96.7 | —Observed 2025-04-16 | — | Source |
| AIME 2025 | 88.9 | AIME 2025 pass@1, high reasoningObserved 2025-04-16 | — | Source |
| Humanity's Last Exam | 20.3 | Humanity's Last Exam, no tools, high reasoningObserved 2025-04-16 | — | Source |
Migration checks
No linked migration route is available for this model yet.
Rankings & picks(4)
Compare o3 with other models
Comparison and alternatives
Browse all comparisons →Show all 53 popular comparisonssorted by 7-day search impressions
Frequently asked questions
What is the context window of o3?
o3 has a context window of 200k tokens.
What is the max output of o3?
o3 can generate up to 100,000 output tokens.
How much does o3 cost?
o3 pricing ranges from $2.00/1M to $2/1M input tokens depending on the provider.
When was o3 released?
o3 was released on 2025-04-16.
Which providers offer o3?
o3 is available from 3 providers: OpenAI API, OpenRouter, Vercel AI Gateway.
What benchmarks has o3 been tested on?
o3 has been evaluated on 12 benchmarks, including HumanEval, SWE-bench Verified, LiveCodeBench, Aider Polyglot, Chatbot Arena.
Cheapest of 3 routes · OpenAI API · cache read $0.500