How many open weight LLM models are available?

There are currently 94 open weight models tracked on Sector HQ. Pricing is collected every 6 hours from all major inference providers.

How is blended price calculated?

Blended price uses the formula (3 × input + 1 × output) / 4 per million tokens, reflecting typical usage where input tokens outnumber output tokens roughly 3:1.

Cheapest Open-Weight / Open Source LLM APIs in 2026

Open-weight LLMs you can self-host or access through inference providers. Compare pricing for Llama, Mistral, Qwen, and other open models.

What is the cheapest open-weight / open source LLM?

The cheapest open-weight / open source LLM API is Meta: Llama 3.2 1B Instruct via Deepinfra at $0.0063 per million tokens. 94 open weight models are compared across all providers.

Open-Weight / Open Source Models by Price

94 models

#	Model	Provider	Input	Output	Quality	Released↓
38	DeepSeek: DeepSeek V3.2 Exp Open	Deepinfra	$0.210	$0.320	Frontier	Dec 1
79	Mistral: Devstral 2 2512 Open	OpenRouter*	$0.480	$2.40	—	Dec 1
62	Qwen: Qwen-Plus Open	OpenRouter*	$0.312	$0.936	—	Jul 28
63	Qwen: Qwen Plus 0728 (thinking) Open	OpenRouter*	$0.312	$0.936	—	Jul 28
14	Qwen: Qwen3 Coder 30B A3B Instruct 30BOpen	Deepinfra	$0.070	$0.260	Good	Jul 14
55	Qwen: Qwen3 Coder Next Open	OpenRouter*	$0.180	$0.960	Good	Jul 14
61	Qwen: Qwen3 Coder Flash Open	OpenRouter*	$0.234	$1.17	Good	Jul 14
65	Qwen: Qwen3 Coder 480B A35B 480BOpen	OpenRouter*	$0.264	$1.20	Good	Jul 14
71	Qwen: Qwen3 Coder 480B A35B (exacto) 480BOpen	OpenRouter*	$0.264	$2.16	Good	Jul 14
85	Qwen: Qwen3 Coder Plus Open	OpenRouter*	$0.780	$3.90	Good	Jul 14
28	Mistral: Voxtral Small 24B 2507 24BOpen	OpenRouter*	$0.120	$0.360	—	Jul 1
80	Mistral: Devstral Medium Open	OpenRouter*	$0.480	$2.40	—	Jul 1
25	Mistral: Devstral Small 1.1 Open	OpenRouter*	$0.120	$0.360	—	May 29
51	DeepSeek: DeepSeek V3.1 Terminus Open	Deepinfra	$0.210	$0.790	Frontier	May 28
59	DeepSeek: DeepSeek V3.1 Terminus (exacto) Open	OpenRouter*	$0.252	$0.948	Frontier	May 28
77	DeepSeek: R1 0528 Open	Deepinfra	$0.500	$2.15	Frontier	May 28
13	Qwen: Qwen3 235B A22B Instruct 2507 235BOpen	OpenRouter*	$0.085	$0.120	Good	Apr 28
16	Qwen: Qwen3 14B 14BOpen	OpenRouter*	$0.072	$0.288	Good	Apr 28
18	Qwen: Qwen3 30B A3B 30BOpen	Deepinfra	$0.080	$0.280	Good	Apr 28
19	Qwen: Qwen3 32B 32BOpen	Deepinfra	$0.080	$0.280	Good	Apr 28
21	Qwen: Qwen3 8B 8BOpen	OpenRouter*	$0.060	$0.480	Good	Apr 28
22	Qwen: Qwen3 30B A3B Instruct 2507 30BOpen	OpenRouter*	$0.108	$0.360	Good	Apr 28
29	Qwen: Qwen3 VL 8B Instruct 8BOpen	Novita AI	$0.080	$0.500	Good	Apr 28
31	Qwen: Qwen3 30B A3B Thinking 2507 30BOpen	OpenRouter*	$0.096	$0.480	Good	Apr 28
35	Qwen: Qwen3 VL 32B Instruct 32BOpen	OpenRouter*	$0.125	$0.499	Good	Apr 28
43	Qwen: Qwen3 VL 30B A3B Instruct 30BOpen	Deepinfra	$0.150	$0.600	Good	Apr 28
45	Qwen: Qwen3 235B A22B 235BOpen	Deepinfra	$0.180	$0.540	Good	Apr 28
49	Qwen: Qwen3 Next 80B A3B Thinking 80BOpen	OpenRouter*	$0.117	$0.936	Good	Apr 28
50	Qwen: Qwen3 Next 80B A3B Instruct 80BOpen	Deepinfra	$0.090	$1.10	Good	Apr 28
54	Qwen: Qwen3 VL 235B A22B Instruct 235BOpen	Deepinfra	$0.200	$0.880	Good	Apr 28
56	Qwen: Qwen3 235B A22B Thinking 2507 235BOpen	Fireworks AI	$0.220	$0.880	Good	Apr 28
57	Qwen: Qwen3 VL 235B A22B Thinking 235BOpen	Fireworks AI	$0.220	$0.880	Good	Apr 28
58	Qwen: Qwen3 VL 30B A3B Thinking 30BOpen	Novita AI	$0.200	$1.00	Good	Apr 28
66	Qwen: Qwen3 VL 8B Thinking 8BOpen	OpenRouter*	$0.140	$1.64	Good	Apr 28
69	Qwen: Qwen3.5 Plus 2026-02-15 Open	OpenRouter*	$0.312	$1.87	Good	Apr 28
82	Qwen: Qwen3.5 397B A17B 397BOpen	OpenRouter*	$0.468	$2.81	Good	Apr 28
87	Qwen: Qwen3 Max Open	OpenRouter*	$0.936	$4.68	Good	Apr 28
88	Qwen: Qwen3 Max Thinking Open	OpenRouter*	$0.936	$4.68	Good	Apr 28
34	Qwen: QwQ 32B 32BOpen	Deepinfra	$0.150	$0.400	—	Mar 6
53	Mistral: Saba Open	OpenRouter*	$0.240	$0.720	—	Feb 1
44	DeepSeek: R1 Distill Qwen 32B 32BOpen	Deepinfra	$0.270	$0.270	Frontier	Jan 20
70	DeepSeek: R1 Distill Llama 70B 70BOpen	Deepinfra	$0.700	$0.800	Frontier	Jan 20
23	NVIDIA: Llama 3.3 Nemotron Super 49B V1.5 49BOpen	Deepinfra	$0.100	$0.400	—	Dec 6
30	Meta: Llama 3.3 70B Instruct 70BOpen	OpenRouter*	$0.120	$0.384	Strong	Dec 6
74	Qwen2.5 Coder 32B Instruct 32BOpen	OpenRouter*	$0.792	$1.20	—	Nov 12
15	Mistral: Ministral 3 3B 2512 3BOpen	OpenRouter*	$0.120	$0.120	—	Oct 16
27	Mistral: Ministral 3 8B 2512 8BOpen	OpenRouter*	$0.180	$0.180	—	Oct 16
40	Mistral: Ministral 3 14B 2512 14BOpen	OpenRouter*	$0.240	$0.240	—	Oct 16
76	NVIDIA: Llama 3.1 Nemotron 70B Instruct 70BOpen	Fireworks AI	$0.900	$0.900	—	Oct 1
1	Meta: Llama 3.2 1B Instruct 1BOpen	Deepinfra	$0.0050	$0.010	Budget	Sep 25

* Prices include ~20% OpenRouter platform fee (verified against actual billing).

All Models Frontier Chat Embedding

Frequently Asked Questions

The cheapest open-weight / open source LLM API is Meta: Llama 3.2 1B Instruct via Deepinfra at $0.0063 per million tokens (blended rate). There are 94 open weight models available.