High-volume, cost-sensitive work
Choose GPT-4.1
GPT-4.1 costs 13% less on a typical mix of three input tokens for every output token ($3.50 against $4.00 per million tokens).
Choose GPT-4.1 for high-volume, cost-sensitive work and writing and code generation. Choose Kimi K3.1 Pro for reused prompts and documents (RAG, agents) and the latest release.
Each answer below comes straight from the published prices, specifications and independent benchmark results.
Choose GPT-4.1
GPT-4.1 costs 13% less on a typical mix of three input tokens for every output token ($3.50 against $4.00 per million tokens).
Choose GPT-4.1
GPT-4.1 charges $8.00 per million output tokens, against $10.00 for Kimi K3.1 Pro. Output dominates the bill when answers are long.
Choose Kimi K3.1 Pro
Kimi K3.1 Pro charges $0.20 per million cached input tokens, against $0.50 for GPT-4.1.
Choose Kimi K3.1 Pro
Kimi K3.1 Pro is the newer release (Oct 23, 2026, against Apr 14, 2025 for GPT-4.1).
| Volume | GPT-4.1 | Kimi K3.1 Pro | Difference |
|---|---|---|---|
| Prototype: 1M input, 250K output tokens | $4.00 | $4.50 | GPT-4.1, $0.50 less |
| Production: 20M input, 5M output tokens | $80.00 | $90.00 | GPT-4.1, $10.00 less |
| Scale: 500M input, 100M output tokens | $1,800 | $2,000 | GPT-4.1, $200.00 less |
Standard rates, without caching or batch discounts. Volumes are per month.
| GPT-4.1 | Kimi K3.1 Pro | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Category | Language models | Language models |
| Status | Deprecated | Generally available |
| Headline price | $3.50 / 1M tokens | $4.00 / 1M tokens |
| Input | $2.00 / 1M tokens | $2.00 / 1M tokens |
| Output | $8.00 / 1M tokens | $10.00 / 1M tokens |
| Cached input | $0.50 / 1M tokens | $0.20 / 1M tokens |
| Context window | 1.05M | 1.05M |
| Max output | 33k | 944k |
| Released | Apr 14, 2025 | Oct 23, 2026 |
| Knowledge cutoff | June 2024 | No data |
| Licence | Proprietary | Proprietary |
| Capabilities | Vision, Tool calling, Structured output, Prompt caching, Batch API | Vision, Reasoning, Tool calling, Structured output, Prompt caching |
| API model id | gpt-4.1 | kimi-k3.1-pro |
Leader in bold.
| Benchmark | GPT-4.1 | Kimi K3.1 Pro |
|---|---|---|
| Terminal-Bench Hard | 13.6% | No data |
| GPQA Diamond | 66.6% | No data |
| Humanity's Last Exam | 4.2% | No data |
| IFBench (instruction following) | 43% | No data |
| τ²-Bench (agentic tool use) | 47.1% | No data |
| MMMU-Pro (visual reasoning) | 61.2% | No data |
| Long-context reasoning (AA-LCR) | 68.3% | No data |
Independent evaluations by Artificial Analysis, run the same way for every model, at the reasoning setting shown. Percentages unless marked as an index.
Choose GPT-4.1 for high-volume, cost-sensitive work and writing and code generation. Choose Kimi K3.1 Pro for reused prompts and documents (RAG, agents) and the latest release.
GPT-4.1 is 13% cheaper than Kimi K3.1 Pro at the headline price ($3.50 / 1M tokens vs $4.00 / 1M tokens).
Kimi K3.1 Pro reads more per request: 1.05M tokens against 1.05M.
GPT-4.1 is made by OpenAI and Kimi K3.1 Pro by Moonshot AI. GPT-4.1 is 13% cheaper than Kimi K3.1 Pro at the headline price ($3.50 / 1M tokens vs $4.00 / 1M tokens). Kimi K3.1 Pro reads more per request: 1.05M tokens against 1.05M. Kimi K3.1 Pro is the newer model (Oct 23, 2026, against Apr 14, 2025). Only GPT-4.1 offers batch api. Only Kimi K3.1 Pro offers reasoning.
At 20 million input and 5 million output tokens a month, GPT-4.1 costs $80.00 and Kimi K3.1 Pro costs $90.00, at standard rates.
Kimi K3.1 Pro was released on Oct 23, 2026 and GPT-4.1 on Apr 14, 2025.