In brief: This historical snapshot lists the per-token rates recorded from the FlowBarAI pricing API on September 12, 2026 — 104 of 126 catalog entries — sorted by input price, lowest first. Input rates run from $0 to $142.5 per 1M tokens, output rates from $0 to $570. Prices were read on September 12, 2026 and are subject to change.
How to read this table
This dated comparison illustrates how to compare input and output costs. It does not quote current rates or guarantee that a listed model remains available.
- All rates are USD per 1,000,000 tokens, captured on September 12, 2026. Consult the pricing page before use.
- 104 of the 126 catalog entries publish both an input and an output rate per 1M tokens; the remaining 22 are image and video generation models billed per generated image or second, which a per-token rate cannot express.
- Legacy membership-gate columns have been removed. Historical prices do not establish current availability.
- Click a column header in the table to re-sort it. The default order is input price, lowest first.
All 104 per-token rates, sorted by input price
| # | Model | Vendor | Input $/1M | Output $/1M | Output ÷ Input |
|---|---|---|---|---|---|
| 1 | deepseek-ai/DeepSeek-OCR | deepseek | $0 | $0 | — |
| 2 | flowbar-free | flowbar | $0 | $0 | — |
| 3 | Qwen/Qwen3-Coder-30B-A3B-Instruct | qwen | $0.0665 | $0.266 | 4× |
| 4 | Qwen/Qwen3-32B | qwen | $0.076 | $0.266 | 3.5× |
| 5 | text-embedding-v3 | qwen | $0.0826 | $0 | 0× |
| 6 | text-embedding-v4 | qwen | $0.0826 | $0 | 0× |
| 7 | deepseek-v4-flash-0731 | deepseek | $0.0855 | $0.171 | 2× |
| 8 | deepseek-v4-flash-0731-new | deepseek | $0.0855 | $0.171 | 2× |
| 9 | openai/gpt-4.1-nano | openai | $0.095 | $0.38 | 4× |
| 10 | openai/gpt-5.6-luna | openai | $0.095 | $0.57 | 6× |
| 11 | qwen3-30b-a3b | qwen | $0.114 | $0.475 | 4.16667× |
| 12 | tencent/hy3 | tencent | $0.1254 | $0.5016 | 4× |
| 13 | deepseek-ai/DeepSeek-V4-Flash | deepseek | $0.133 | $0.266 | 2× |
| 14 | deepseek-v4-flash | deepseek | $0.133 | $0.266 | 2× |
| 15 | Qwen3.5-35B-A3B | qwen | $0.133 | $0.95 | 7.14286× |
| 16 | qwen3.5-35b-a3b | qwen | $0.133 | $0.95 | 7.14286× |
| 17 | qwen3.6-35b-a3b | qwen | $0.133 | $0.95 | 7.14286× |
| 18 | qwen3-235b-a22b-instruct-2507 | qwen | $0.142 | $0.5681 | 4× |
| 19 | openai/gpt-4o-mini | openai | $0.1425 | $0.57 | 4× |
| 20 | qwen3-vl-30b-a3b-instruct | qwen | $0.1425 | $0.57 | 4× |
| 21 | minimax-m2.5 | minimax | $0.1425 | $0.855 | 6× |
| 22 | qwen3.6-flash | qwen | $0.1781 | $1.0687 | 6× |
| 23 | meta-llama/llama-4-maverick | meta | $0.19 | $0.76 | 4× |
| 24 | stepfun/step-3.7-flash | stepfun | $0.19 | $1.0925 | 5.75× |
| 25 | qwen3-vl-235b-a22b-instruct | qwen | $0.1995 | $1.805 | 9.04762× |
| 26 | qwen-vl-chat | qwen | $0.2153 | $0.646 | 3.00065× |
| 27 | qwen3-235b-a22b-thinking-2507 | qwen | $0.2185 | $2.185 | 10× |
| 28 | qwen2.5-vl-72b-instruct | qwen | $0.2375 | $0.7125 | 3× |
| 29 | minimax-m2.7 | minimax | $0.2375 | $0.95 | 4× |
| 30 | google/gemini-3.1-flash-lite | $0.2375 | $1.425 | 6× | |
| 31 | openai/gpt-5-mini | openai | $0.2375 | $1.9 | 8× |
| 32 | deepseek-chat | deepseek | $0.2445 | $0.9773 | 3.9965× |
| 33 | DeepSeek-V3.2-Pro | deepseek | $0.2555 | $0.38 | 1.48699× |
| 34 | deepseek-v3.2 | deepseek | $0.2555 | $0.38 | 1.48699× |
| 35 | deepseek-v3.2-1222 | deepseek | $0.2555 | $0.38 | 1.48699× |
| 36 | deepseek-v3.2-standard | deepseek | $0.2555 | $0.38 | 1.48699× |
| 37 | qwen/qwen3-coder | qwen | $0.285 | $0.95 | 3.33333× |
| 38 | qwen3-coder-480b-a35b-instruct | qwen | $0.285 | $0.95 | 3.33333× |
| 39 | minimax-m3 | minimax | $0.285 | $1.14 | 4× |
| 40 | minimax/minimax-m3 | minimax | $0.285 | $1.14 | 4× |
| 41 | Qwen/Qwen3.6-27B | qwen | $0.285 | $1.9 | 6.66667× |
| 42 | google/gemini-2.5-flash | $0.285 | $2.375 | 8.33333× | |
| 43 | qwen/qwen3.7-plus | qwen | $0.304 | $1.216 | 4× |
| 44 | qwen3.6-plus | qwen | $0.3087 | $1.8525 | 6× |
| 45 | Qwen2.5-72B-Instruct | qwen | $0.342 | $0.38 | 1.11111× |
| 46 | Qwen3.5-397B-A17B-Pro | qwen | $0.3705 | $2.223 | 6× |
| 47 | openai/gpt-4.1-mini | openai | $0.38 | $1.52 | 4× |
| 48 | deepseek-v4-pro | deepseek | $0.4133 | $0.8265 | 2× |
| 49 | deepseek-v4-pro-0813 | deepseek | $0.4133 | $0.8265 | 2× |
| 50 | deepseek-v4-pro-0817 | deepseek | $0.4133 | $0.8265 | 2× |
| 51 | openai/gpt-3.5-turbo | openai | $0.475 | $1.425 | 3× |
| 52 | DeepSeek-R1-0528 | deepseek | $0.475 | $2.0425 | 4.3× |
| 53 | Kimi-K2 | moonshot | $0.5415 | $2.185 | 4.03509× |
| 54 | kimi-k2.5 | moonshot | $0.5415 | $2.7075 | 5× |
| 55 | Pro/moonshotai/Kimi-K2.6 | moonshot | $0.5596 | $2.356 | 4.21053× |
| 56 | kimi-k2.6 | moonshot | $0.5596 | $2.356 | 4.21053× |
| 57 | Kimi-K2-Thinking | moonshot | $0.57 | $2.375 | 4.16667× |
| 58 | nvidia/nemotron-3-ultra-550b-a55b | nvidia | $0.57 | $3.42 | 6× |
| 59 | deepseek-reasoner | deepseek | $0.665 | $2.375 | 3.57143× |
| 60 | kwaipilot/kat-coder-pro-v2.5 | kwaipilot | $0.703 | $2.812 | 4× |
| 61 | google/gemini-3.7-flash | $0.7125 | $3.5625 | 5× | |
| 62 | google/gemini-3.8-flash | $0.7125 | $3.5625 | 5× | |
| 63 | glm-5.2 | zai | $0.722 | $2.299 | 3.18421× |
| 64 | glm-5.3 | zai | $0.722 | $2.299 | 3.18421× |
| 65 | z-ai/glm-5.2 | zai | $0.722 | $2.299 | 3.18421× |
| 66 | qwen3-max | qwen | $0.741 | $3.705 | 5× |
| 67 | Pro/zai-org/GLM-5 | zai | $0.9025 | $2.4225 | 2.68421× |
| 68 | glm-5 | zai | $0.9025 | $2.4225 | 2.68421× |
| 69 | Pro/zai-org/GLM-5.1 | zai | $0.9177 | $2.8842 | 3.14286× |
| 70 | glm-5.1 | zai | $0.9177 | $2.8842 | 3.14286× |
| 71 | perplexity/sonar | perplexity | $0.95 | $0.95 | 1× |
| 72 | anthropic/claude-haiku-4.5 | anthropic | $0.95 | $4.75 | 5× |
| 73 | openai/gpt-5.6-terra | openai | $0.95 | $5.7 | 6× |
| 74 | openai/o3-mini | openai | $1.045 | $4.18 | 4× |
| 75 | x-ai/grok-4.20 | xai | $1.1875 | $2.375 | 2× |
| 76 | google/gemini-2.5-pro | $1.1875 | $9.5 | 8× | |
| 77 | openai/gpt-5.1-chat | openai | $1.1875 | $9.5 | 8× |
| 78 | qwen/qwen3.7-max | qwen | $1.4013 | $4.2037 | 3× |
| 79 | qwen3.8-max | qwen | $1.4013 | $4.2037 | 3× |
| 80 | google/gemini-3.5-flash | $1.425 | $8.55 | 6× | |
| 81 | mistralai/mistral-large | mistral | $1.9 | $5.7 | 3× |
| 82 | x-ai/grok-4.5 | xai | $1.9 | $5.7 | 3× |
| 83 | x-ai/grok-4.6 | xai | $1.9 | $5.7 | 3× |
| 84 | anthropic/claude-sonnet-5 | anthropic | $1.9 | $9.5 | 5× |
| 85 | google/gemini-3.1-pro-preview | $1.9 | $11.4 | 6× | |
| 86 | openai/gpt-4o | openai | $2.375 | $9.5 | 4× |
| 87 | kimi-k3 | moonshot | $2.85 | $14.25 | 5× |
| 88 | moonshotai/kimi-k3 | moonshot | $2.85 | $14.25 | 5× |
| 89 | perplexity/sonar-pro-search | perplexity | $2.85 | $14.25 | 5× |
| 90 | anthropic/claude-opus-4.8 | anthropic | $4.75 | $23.75 | 5× |
| 91 | openai/gpt-5.5 | openai | $4.75 | $28.5 | 6× |
| 92 | openai/gpt-5.6-sol | openai | $4.75 | $28.5 | 6× |
| 93 | sakana/fugu-ultra | sakana | $4.75 | $28.5 | 6× |
| 94 | openai/gpt-5-image | openai | $9.5 | $9.5 | 1× |
| 95 | openai/o3-deep-research | openai | $9.5 | $38 | 4× |
| 96 | anthropic/claude-fable-5 | anthropic | $9.5 | $47.5 | 5× |
| 97 | anthropic/claude-fable-5.1 | anthropic | $9.5 | $47.5 | 5× |
| 98 | anthropic/claude-opus-4.8-fast | anthropic | $9.5 | $47.5 | 5× |
| 99 | openai/gpt-6-astra | openai | $9.5 | $47.5 | 5× |
| 100 | openai/gpt-6-astra-pro | openai | $9.5 | $47.5 | 5× |
| 101 | openai/gpt-5-pro | openai | $14.25 | $114 | 8× |
| 102 | openai/o3-pro | openai | $19 | $76 | 4× |
| 103 | openai/gpt-5.5-pro | openai | $28.5 | $171 | 6× |
| 104 | openai/o1-pro | openai | $142.5 | $570 | 4× |
Rates read from FlowBarAI's pricing API on September 12, 2026. Prices are subject to change; the authoritative source is the pricing page.
Balance-based model access
No subscription or membership upgrade is required. FlowBarAI users can use currently available models with sufficient balance. Model availability, account safety checks and rate limits still apply.
What the historical numbers show
- Input rates span $0 to $142.5 per 1M tokens (openai/o1-pro is the most expensive input rate in the catalog).
- Output rates span $0 to $570 per 1M tokens.
- 10 entries are priced at or below $0.10 per 1M input tokens, and 73 at or below $1.00 per 1M input tokens.
- Output costs more than input almost everywhere: the median output rate is 4× the input rate across the 102 entries where both are non-zero.
- The cheapest paid per-token entry is Qwen/Qwen3-Coder-30B-A3B-Instruct at $0.0665 per 1M input tokens and $0.266 per 1M output tokens.
Models billed outside per-token pricing
These 22 entries in the September 12, 2026 snapshot are image and video generation models. They publish no per-1M-token rate, so they are listed here by name instead of being given a price this page cannot support. Per-generation rates are on the pricing page.
- doubao-seedream-4.0
- doubao-seedream-4.5
- Doubao-seedream-5.0-lite
- doubao-seedream-5.0-pro
- happyhorse-1.0-i2v
- happyhorse-1.0-r2v-20260618
- happyhorse-1.0-t2v
- happyhorse-1.1-i2v
- happyhorse-1.1-r2v
- happyhorse-1.1-t2v
- qwen-image-edit-max
- qwen-image-edit-plus
- wan2.2-kf2v-flash
- wan2.6-r2v
- wan2.6-r2v-flash
- wan2.7-i2v
- wan2.7-image
- wan2.7-image-pro
- wan2.7-r2v
- wan2.7-t2v
- wan3.0-video
- wan3.0-video-prime
How to compare these rates with another provider
The same method is written up in more detail in how to compare LLM API pricing, with cost-saving reasoning in how to choose a cheap LLM API. The catalog behind this table is browsable on the models page, and endpoint details are in the docs.
- Convert both sides to the same unit. Per-1,000 and per-1,000,000 quotes are mixed constantly, and a unit mistake silently changes a ranking by 1,000×.
- Split your workload into input tokens and output tokens, then multiply each by its own rate. The output multiple is large enough that a blended number misleads.
- Confirm the model is currently available and that your balance covers its cost.
- Add the costs outside token pricing: cache reads and writes, image and video generation, and any subscription or minimum spend.
- Confirm the failure policy. Failed requests are not charged on FlowBarAI; check each other provider's terms.
Frequently asked questions
Which LLM API is cheapest per token?
In the pricing data read on September 12, 2026, 2 entries are listed at $0 per 1M input and output tokens. Among paid entries the lowest input rate is $0.0665 per 1M tokens (Qwen/Qwen3-Coder-30B-A3B-Instruct), but the cheapest choice still depends on your output volume and current model availability.
Are these prices current?
They were read from FlowBarAI's pricing API on September 12, 2026 and are subject to change. Consult the pricing page for published rates and check current model availability before use.
Why do output rates differ so much from input rates?
In this catalog the median output rate is 4× the input rate across the 102 entries where both are non-zero. Output-heavy workloads are therefore dominated by the output rate, while summarization and classification are dominated by input.
Do I need a membership upgrade to use models?
No subscription or membership upgrade is required. FlowBarAI users can use currently available models with sufficient balance. Model availability, account safety checks and rate limits still apply. Check the published rate for the model you select.
How many models are in the catalog?
126 entries as read on September 12, 2026: 104 with published per-1M-token rates and 22 image or video generation models billed per generated unit.
What happens if a request fails?
Failed requests are not charged on FlowBarAI.
Related reading
- OpenRouter alternatives: multi-model and self-hosted options
- Free LLM API access: what a real free trial includes
- LiteLLM alternatives: self-hosted vs managed
- How to choose a cheap LLM API
- How to compare LLM API pricing
- AI video generation platform workflow
- FlowBarAI in English: gateway overview
Check the live rates
The pricing page carries current per-model rates. Check the models page for current availability.