Comparisons
Claude vs GPT: API price and fit
Claude and GPT API models compared by per-token price, cache rate, endpoints and tooling. Choose per task and switch between them with one OpenAI-compatible key.
tokely.me/claude-vs-gpt
Best LLM for coding: how to choose
Coding models compared by price and fit: Claude Sonnet and Opus, GPT Codex, Kimi K2.7 Code, DeepSeek V4 Pro and Gemini Pro. Test them on one key in your own tools.
By TokelyReviewed There is no single best coding LLM for every team. Claude Sonnet and Opus, GPT Codex models, Kimi K2.7 Code, DeepSeek V4 Pro and Gemini Pro are the common candidates. Pick two by budget, run them on your own tasks in your coding agent, and compare cost per accepted change. All are available on one Tokely key.
Price comparison of the models on this page
| Model | Tokely input / output | Tokely example | model developer example | Discount vs model developer |
|---|---|---|---|---|
| GPT-5.4 MiniOpenAI | $0.3 / $1.8Cache read: $0.03 | $0.75 | $1.875$0.75 / $4.5 per 1M | 60.0% off |
| Claude Haiku 5.5Anthropic | $0.042353 / $0.211764Cache read: $0.042353 | $0.095294 | $0.225$0.1 / $0.5 per 1M | 57.6% off |
| GPT-5 NanoOpenAI | $0.035 / $0.28Cache read: $0.0035 | $0.105 | $0.15$0.05 / $0.4 per 1M | 30.0% off |
| Gemini 2.5 Flash LiteGoogle | $0.07 / $0.28Cache read: $0.007 | $0.14 | $0.2$0.1 / $0.4 per 1M | 30.0% off |
| GPT-4.1 NanoOpenAI | $0.07 / $0.28Cache read: $0.0175 | $0.14 | $0.2$0.1 / $0.4 per 1M | 30.0% off |
| GPT-6 LunaOpenAI | $0.07 / $0.35Cache read: $0.007 | $0.1575 | $0.225$0.1 / $0.5 per 1M | 30.0% off |
| GPT-4o MiniOpenAI | $0.105 / $0.42Cache read: $0.0525 | $0.21 | $0.3$0.15 / $0.6 per 1M | 30.0% off |
| GPT-OSS 120BOpenAI | $0.1331 / $0.5324Cache read: $0.1331 | $0.2662 | Not matched | — |
| Gemini Embedding 001Google | $0.158825 / $0.635299Cache read: $0.158825 | $0.31765 | Not matched | — |
| GPT-5.6 LunaOpenAI | $0.14 / $0.84Cache read: $0.014 | $0.35 | $0.5$0.2 / $1.2 per 1M | 30.0% off |
| DeepSeek V4.1 FlashDeepSeek | $0.2 / $0.8Cache read: $0.004 | $0.4 | $0.3$0.15 / $0.6 per 1M | 33.3% higher |
| Gemini 3.1 Flash LiteGoogle | $0.175 / $1.05Cache read: $0.0175 | $0.4375 | $0.625$0.25 / $1.5 per 1M | 30.0% off |
| GPT-5 MiniOpenAI | $0.175 / $1.4Cache read: $0.0175 | $0.525 | $0.75$0.25 / $2 per 1M | 30.0% off |
| GPT-5.1 Codex MiniOpenAI | $0.175 / $1.4Cache read: $0.0175 | $0.525 | $0.75$0.25 / $2 per 1M | 30.0% off |
| GPT-4.1 MiniOpenAI | $0.28 / $1.12Cache read: $0.07 | $0.56 | $0.8$0.4 / $1.6 per 1M | 30.0% off |
| GPT-5.4 NanoOpenAI | $0.242 / $1.5125Cache read: $0.242 | $0.620125 | $0.5125$0.2 / $1.25 per 1M | 21.0% higher |
| Gemini 2.5 FlashGoogle | $0.21 / $1.75Cache read: $0.021 | $0.6475 | $0.925$0.3 / $2.5 per 1M | 30.0% off |
| Gemini 3.5 Flash LiteGoogle | $0.21 / $1.75Cache read: $0.021 | $0.6475 | $0.925$0.3 / $2.5 per 1M | 30.0% off |
| DeepSeek V3.1DeepSeek | $0.478335 / $1.435005Cache read: $0.478335 | $0.837086 | Not matched | — |
| Gemini 3 Flash PreviewGoogle | $0.35 / $2.1Cache read: $0.035 | $0.875 | $1.25$0.5 / $3 per 1M | 30.0% off |
| GPT-3.5 TurboOpenAI | $0.528 / $1.584Cache read: $0.528 | $0.924 | $0.875$0.5 / $1.5 per 1M | 5.6% higher |
| DeepSeek V4 FlashDeepSeek | $0.605 / $1.815Cache read: $0.01925 | $1.05875 | Not matched | — |
| Gemini 3.7 FlashGoogle | $0.525 / $2.625Cache read: $0.0525 | $1.18125 | $1.6875$0.75 / $3.75 per 1M | 30.0% off |
| Gemini 3.8 FlashGoogle | $0.525 / $2.625Cache read: $0.0525 | $1.18125 | $1.6875$0.75 / $3.75 per 1M | 30.0% off |
| Gemini 3.6 FlashGoogle | $0.525 / $2.625Cache read: $0.0525 | $1.18125 | $1.6875$0.75 / $3.75 per 1M | 30.0% off |
| o3 MiniOpenAI | $0.77 / $3.08Cache read: $0.385 | $1.54 | $2.2$1.1 / $4.4 per 1M | 30.0% off |
| o4 MiniOpenAI | $0.77 / $3.08Cache read: $0.1925 | $1.54 | $2.2$1.1 / $4.4 per 1M | 30.0% off |
| GPT-6.1 SolOpenAI | $0.8 / $4Cache read: $0.04 | $1.8 | $4.5$2 / $10 per 1M | 60.0% off |
| Kimi K2.5Moonshot | $0.825 / $4.33125Cache read: $0.825 | $1.907813 | Not matched | — |
| Gemini 2.5 ProGoogle | $0.875 / $7Cache read: $0.0875 | $2.625 | $3.75$1.25 / $10 per 1M | 30.0% off |
| Gemini 3.5 FlashGoogle | $1.05 / $6.3Cache read: $0.105 | $2.625 | $3.75$1.5 / $9 per 1M | 30.0% off |
| GPT-5 CodexOpenAI | $0.875 / $7Cache read: $0.0875 | $2.625 | $3.75$1.25 / $10 per 1M | 30.0% off |
| GPT-5.1 CodexOpenAI | $0.875 / $7Cache read: $0.0875 | $2.625 | $3.75$1.25 / $10 per 1M | 30.0% off |
| GPT-5OpenAI | $0.875 / $7Cache read: $0.0875 | $2.625 | $3.75$1.25 / $10 per 1M | 30.0% off |
| GPT-5.1OpenAI | $0.875 / $7Cache read: $0.0875 | $2.625 | $3.75$1.25 / $10 per 1M | 30.0% off |
| Kimi K2.6Moonshot | $1.30625 / $5.426025Cache read: $0.221058 | $2.662756 | $1.95$0.95 / $4 per 1M | 36.6% higher |
| Kimi K2.7 CodeMoonshot | $1.30625 / $5.426025Cache read: $0.26125 | $2.662756 | $1.95$0.95 / $4 per 1M | 36.6% higher |
| GPT-4.1OpenAI | $1.4 / $5.6Cache read: $0.35 | $2.8 | $4$2 / $8 per 1M | 30.0% off |
| o3OpenAI | $1.4 / $5.6Cache read: $0.35 | $2.8 | $4$2 / $8 per 1M | 30.0% off |
| GPT-6 SolOpenAI | $1.4 / $7Cache read: $0.14 | $3.15 | $4.5$2 / $10 per 1M | 30.0% off |
| Gemini 4 ArgonGoogle | $1.4 / $7Cache read: $0.07 | $3.15 | $4.5$2 / $10 per 1M | 30.0% off |
| DeepSeek V4 ProDeepSeek | $1.815 / $5.445Cache read: $0.060494 | $3.17625 | Not matched | — |
| Claude Sonnet 5Anthropic | $1.5 / $7.5Cache read: $0.15 | $3.375 | $4.5$2 / $10 per 1M | 25.0% off |
| Claude Sonnet 5.5Anthropic | $1.5 / $7.5Cache read: $0.075 | $3.375 | $4.5$2 / $10 per 1M | 25.0% off |
| GPT-Chat LatestOpenAI | $1.375 / $8.25Cache read: $1.375 | $3.4375 | Not matched | — |
| GPT-5.6 TerraOpenAI | $1.4 / $8.4Cache read: $0.14 | $3.5 | $5$2 / $12 per 1M | 30.0% off |
| Gemini 3.1 Pro PreviewGoogle | $1.4 / $8.4Cache read: $0.14 | $3.5 | $5$2 / $12 per 1M | 30.0% off |
| GPT-4oOpenAI | $1.75 / $7Cache read: $0.875 | $3.5 | $5$2.5 / $10 per 1M | 30.0% off |
| GPT-5.2OpenAI | $1.225 / $9.8Cache read: $0.1225 | $3.675 | $5.25$1.75 / $14 per 1M | 30.0% off |
| GPT-5.2 CodexOpenAI | $1.225 / $9.8Cache read: $0.1225 | $3.675 | $5.25$1.75 / $14 per 1M | 30.0% off |
| GPT-5.3 CodexOpenAI | $1.225 / $9.8Cache read: $0.1225 | $3.675 | $5.25$1.75 / $14 per 1M | 30.0% off |
| GPT-5.4OpenAI | $1.75 / $10.5Cache read: $0.175 | $4.375 | $6.25$2.5 / $15 per 1M | 30.0% off |
| GPT-5.5OpenAI | $2 / $12Cache read: $0.2 | $5 | $12.5$5 / $30 per 1M | 60.0% off |
| Claude Sonnet 4.6Anthropic | $2.25 / $11.25Cache read: $0.225 | $5.0625 | $6.75$3 / $15 per 1M | 25.0% off |
| Kimi K3Moonshot | $2.7 / $13.5Cache read: $0.27 | $6.075 | $6.75$3 / $15 per 1M | 10.0% off |
| GPT-5.6 SolOpenAI | $2.8 / $14Cache read: $0.28 | $6.3 | $9$4 / $20 per 1M | 30.0% off |
| Claude Opus 5.5Anthropic | $3 / $15Cache read: $0.15 | $6.75 | $9$4 / $20 per 1M | 25.0% off |
| Claude Opus 4.6Anthropic | $3.75 / $18.75Cache read: $0.375 | $8.4375 | $11.25$5 / $25 per 1M | 25.0% off |
| Claude Opus 4.7Anthropic | $3.75 / $18.75Cache read: $0.375 | $8.4375 | $11.25$5 / $25 per 1M | 25.0% off |
| Claude Opus 4.8Anthropic | $3.75 / $18.75Cache read: $0.375 | $8.4375 | $11.25$5 / $25 per 1M | 25.0% off |
| Claude Opus 5Anthropic | $3.75 / $18.75Cache read: $0.375 | $8.4375 | $11.25$5 / $25 per 1M | 25.0% off |
| Claude Opus 4.5Anthropic | $3.75 / $18.75Cache read: $0.375 | $8.4375 | $11.25$5 / $25 per 1M | 25.0% off |
| GPT-6 AstraOpenAI | $4 / $20Cache read: $0.4 | $9 | $22.5$10 / $50 per 1M | 60.0% off |
| GPT-4 TurboOpenAI | $7 / $21Cache read: $4.4 | $12.25 | $17.5$10 / $30 per 1M | 30.0% off |
| Claude Fable 5Anthropic | $7.5 / $37.5Cache read: $0.75 | $16.875 | $22.5$10 / $50 per 1M | 25.0% off |
| Claude Fable 5.1Anthropic | $7.5 / $37.5Cache read: $0.1875 | $16.875 | $22.5$10 / $50 per 1M | 25.0% off |
| o1OpenAI | $10.5 / $42Cache read: $5.25 | $21 | $30$15 / $60 per 1M | 30.0% off |
| o3 ProOpenAI | $14 / $56Cache read: $8.8 | $28 | $40$20 / $80 per 1M | 30.0% off |
| GPT-4OpenAI | $21 / $42Cache read: $6.6 | $31.5 | $45$30 / $60 per 1M | 30.0% off |
| GPT-5 ProOpenAI | $10.5 / $84Cache read: $6.6 | $31.5 | $45$15 / $120 per 1M | 30.0% off |
| GPT-5.2 ProOpenAI | $14.7 / $117.6Cache read: $21 | $44.1 | $63$21 / $168 per 1M | 30.0% off |
| GPT-5.4 ProOpenAI | $21 / $126Cache read: $13.2 | $52.5 | $75$30 / $180 per 1M | 30.0% off |
| GPT-5.5 ProOpenAI | $121.3245 / $545.96025Cache read: $0 | $257.814563 | $75$30 / $180 per 1M | 243.8% higher |
Shortlist by budget
Premium: Claude Opus and GPT pro or Codex variants. Mid-range: Claude Sonnet and Gemini Pro. Low-cost: Kimi K2.7 Code and DeepSeek V4 Pro, which are priced well below the premium tier per token.
Agents are dominated by input tokens
Coding agents resend the repository context and tool definitions on every turn, so input and cache-read prices matter more than output price. Prefer models with prompt caching and check cache-read rates in the table.
Measure on your codebase
Give each candidate the same 10 to 20 real tickets in your agent, count accepted changes and total cost, and keep the cheapest model that meets your bar. Use the expensive model only for the hardest tasks.
One key for every model
Every model on this page is available through one OpenAI-compatible endpoint and one balance, with routing modes, bounded fallback, spending limits and request logs.
LLM gateway: one API for every model ↗
A hosted LLM gateway with one OpenAI-compatible endpoint, account routing modes, bounded fallback, spending limits and request logs. Text, image, video and audio from one balance.
OpenAI-compatible API for Claude, Gemini, GPT and more ↗
Keep the OpenAI SDK and call Claude, Gemini, GPT, Kimi, GLM and other supported models. Change the base URL, key and model ID.
Questions before you switch
What is the cheapest good coding model?
Kimi K2.7 Code and DeepSeek V4 Pro are the low-cost candidates on Tokely; whether they are good enough depends on your tasks, so test them against a premium model.
Can I use these models in Cline, Roo or Claude Code?
Yes. OpenAI-compatible agents connect to https://api.tokely.me/v1; Claude Code uses the native /v1/messages endpoint for Claude models. See the coding agents guide.
START WITH YOUR OWN WORKLOADKnow the price.
Know the price.
Then make the call.
Check the model, compare the rate, and connect with a Tokely key.