Price per 1M logo

Cheapest reasoning models

Models flagged as reasoning-capable, sorted by input cost, for maths, planning and multi-step debugging.

Current leader: Gemma 4 26b A4b It (Google Gemini). Refreshed 2026-09-13 13:40 UTC.

#ModelProviderInput $/1MOutput $/1MCached $/1MContext
1Gemma 4 26b A4b ItGoogle Geminifreefree262K
2Gemma 4 31b ItGoogle Geminifreefree262K
3Anthropic.claude Mythos PreviewAmazon Bedrockfreefree1M
4AutoOpenRouterfreefree2M
5FreeOpenRouterfreefree200K
6Nemotron 3.5 Lightning:freeOpenRouterfreefree1M
7Laguna S 2.1:freeOpenRouterfreefree262K
8Laguna Xs 2.1:freeOpenRouterfreefree262K
9Glm 5.2:freeOpenRouterfreefree256K
10Nemotron 3.5 Content Safety:freeOpenRouterfreefree128K
11Nemotron 3 Ultra 550b A55b:freeOpenRouterfreefree1M
12Minimax M3:freeOpenRouterfreefree1.0M
13Nemotron 3 Nano Omni 30b A3b Reasoning:freeOpenRouterfreefree256K
14Gemma 4 26b A4b It:freeOpenRouterfreefree262K
15Gemma 4 31b It:freeOpenRouterfreefree262K
16Minimax M2.7:freeOpenRouterfreefree197K
17Nemotron 3 Super 120b A12b:freeOpenRouterfreefree262K
18Nemotron 3.5 Lightning 30b A3bPerplexity$0.011$0.17$0.0011
19Gemma 4 E4B ItDeepInfra$0.02$0.1131K
20Qwen3 4b Fp8Novita AI$0.03$0.03128K
21GPT Oss 20bOpenRouter$0.03$0.13131K
22Qwen3.7 FlashOpenRouter$0.03$0.13$0.0061M
23Qwen3 8b Fp8Novita AI$0.035$0.138128K
24GPT Oss 120bOpenRouter$0.037$0.17131K
25GPT Oss 20bNovita AI$0.04$0.15131K
26Gemma 4 26b A4b ItOpenRouter$0.042$0.22262K
27Qwen TurboAlibaba DashScope$0.05$0.2129K
28Qwen TurboAlibaba DashScope$0.05$0.21M
29Qwen TurboAlibaba DashScope$0.05$0.21M
30Qwen TurboAlibaba DashScope$0.05$0.21M
31GPT 5 NanoOpenAI$0.05$0.4$0.005272K
32GPT 5 NanoOpenAI$0.05$0.4$0.005272K
33GPT 5 NanoAzure OpenAI$0.05$0.4$0.005272K
34Nemotron 3 Nano 30B A3BDeepInfra$0.05$0.2$0.025262K
35Nemotron Lightning 3p5 30b A3bFireworks AI$0.05$0.2$0.01262K
36GPT Oss 120bNovita AI$0.05$0.25131K
37Nemotron 3 Nano 30b A3bNovita AI$0.05$0.2262K
38GPT 5 NanoOpenRouter$0.05$0.4$0.005272K
39Nemotron 3 Nano 30b A3bOpenRouter$0.05$0.2$0.03262K
40Qwen3 30b A3b Fp8Cloudflare Workers AI$0.051$0.33533K
41GPT 5 NanoAzure OpenAI$0.055$0.44$0.0055272K
42GLM 4.7 FlashDeepInfra$0.06$0.4$0.01203K
43Ling 3.0 FlashDeepInfra$0.06$0.18$0.012131K
44NVIDIA Nemotron 3 Nano 30B A3BNebius$0.06$0.24262K
45Nemotron 3 Nano OmniNebius$0.06$0.24262K
46Nemotron 3 5 LightningNebius$0.06$0.241.0M
47Deepseek R1 0528 Qwen3 8bNovita AI$0.06$0.09128K
48Ling 3.0 Flash FastNovita AI$0.06$0.18$0.012262K
49Ling 3.0 FlashNovita AI$0.06$0.18$0.012262K
50Glm 4.7 FlashOpenRouter$0.06$0.4$0.01200K
51Laguna Xs 2.1OpenRouter$0.06$0.12$0.03262K
52Glm 4.7 FlashCloudflare Workers AI$0.06$0.4131K
53Qwen3.5 Flash 02 23OpenRouter$0.065$0.261M
54Deepseek V4 Flash 0731OpenRouter$0.065$0.18$0.0161.3M
55Gemma 4 26B A4B ItDeepInfra$0.07$0.34262K
56GPT Oss 20bFireworks AI$0.07$0.3$0.035131K
57Ernie 4.5 21B A3b ThinkingNovita AI$0.07$0.28131K
58Glm 4.7 FlashNovita AI$0.07$0.4$0.01200K
59GPT Oss 20bGroq$0.075$0.3$0.037131K
60GPT Oss Safeguard 20bGroq$0.075$0.3$0.037131K

Other use cases

Frequently asked

Which reasoning model is the cheapest?

Gemma 4 26b A4b It from Google Gemini currently leads this list, at $0/1M input tokens. Rankings are recomputed from the raw dataset on every refresh, so they track price cuts automatically.

How are these rankings calculated?

Each use case applies a capability filter (context size, vision, tool calling, cache pricing) and then ranks the surviving models by the blended cost that profile actually pays. See the methodology page for the exact formulas.

Is the cheapest model always the right choice?

No. Check the capability flags and context window, and benchmark quality on your own data — the cheapest model that fails your evals costs more than the second cheapest that passes them.