Reasoning Models in 2026

Updated 6 October 2026

332 on record. Sorted by context; every figure comes from the OpenRouter model catalogue, with the page it was read from on each entry.

332listed

#	Model	Context window	Input / 1M	Output / 1M
301	Granite 4.2 8BIBMopen weightstext	131K	$0.061	$0.251
302	Hermes 4 405BNousopen weightstext	131K	$11	$31
303	Hunyuan A13B InstructTencentopen weightstext	131K	$0.141	$0.571
304	Muse Glimmer 30BMetaopen weightstext, image	131K	$0.351	$1.51
305	Nano Banana 2 (Gemini 3.1 Flash Image)Googleimage, text	131K	$0.51	$31
306	Nano Banana Pro (Gemini 3 Pro Image)Googleimage, text	131K	$21	$121
307	Nemotron 3.5 Content SafetyNVIDIAopen weightstext, image	131K	$0.21	$0.21
308	Qwen3 14BQwenopen weightstext	131K	$0.121	$0.241
309	Qwen3 235B A22BQwenopen weightstext	131K	$0.455	$1.821
310	Qwen3 235B A22B Thinking 2507Qwenopen weightstext	131K	$0.231	$2.31
311	Qwen3 30B A3BQwenopen weightstext	131K	$0.121	$0.51
312	Qwen3 32BQwenopen weightstext	131K	$0.081	$0.281
313	Qwen3 8BQwenopen weightstext	131K	$0.117	$0.455
314	Qwen3 VL 235B A22B ThinkingQwenopen weightstext, image	131K	$0.41	$41
315	Qwen3 VL 8B ThinkingQwenopen weightsimage, text	131K	$0.181	$2.11
316	Solar Pro 3Upstagetext	131K	$0.151	$0.61
317	Mercury 2Inceptiontext	128K	$0.251	$0.751
318	Nemotron 3.5 Content Safety (free)NVIDIAopen weightstext, image	128K	free	free
319	Sonar Deep ResearchPerplexitytext	128K	$21	$81
320	Sonar Reasoning ProPerplexitytext, image	128K	$21	$81
321	ERNIE 4.5 VL 424B A47B Baiduopen weightsimage, text	123K	$0.421	$1.251
322	Qwen3 30B A3B Thinking 2507Qwenopen weightstext	82K	$0.21	$2.41
323	GLM 4.5VZ.aiopen weightstext, image	66K	$0.61	$1.81
324	LFM2.5-2.6B (free)LiquidAIopen weightstext	66K	free	free
325	Nano Banana 2 (Gemini 3.1 Flash Image Preview)Googleimage, text	66K	$0.51	$31

Context bars share one log scale, 1K to 2M tokens. Prices are what each provider publishes per million tokens; 'varies' is a router that bills the model it picks.

More in Models

All models lists

Cheapest language models 437 Longest-context models 466 Open-weight models 178 Vision language models 295 Free models 22 Newest models 466 Models with tool calling 395 Image generation models 9