Providers
A spread is over distinct sellers, not a vendor’s own service tiers. Where serving precision differs, the cheap row is not a like-for-like saving.
Direct provider price observations
652 source quotes across 3 providers. Check the original rate card, delivery type and observation date before using a quote.
These provider IDs have not been matched to benchmark identities or tested as equivalent endpoints. They are a separate research dataset. EUR observations stay in their source currency and do not enter the USD leaderboard.
Download all source quotes · JSONRead the extraction method
NanoGPT ↗
Observed 2026-10-07 · 612 provider IDs · 612 quotes
Unauthenticated website-host text-model catalogue. IDs can include aliases and routing choices. Base token quotes only; scheduled prices are withheld.
26 IDs excluded by the documented scope checks. Rates below are alphabetical samples, per million tokens.
| Provider ID / delivery | Input /M | Output /M |
|---|---|---|
abacusai/Dracarys-72B-Instructbase · precision unlisted | USD 0.493 | USD 0.493 |
abliteration-ai/abliterated-modelbase · precision unlisted | USD 1.00 | USD 3.00 |
abliteration-ai/abliterated-model-largebase · precision unlisted | USD 3.00 | USD 5.00 |
abliteration-ai/abliterated-model-large-v2base · precision unlisted | USD 3.00 | USD 5.00 |
Chutes ↗
Observed 2026-10-07 · 14 provider IDs · 14 quotes
Public inference catalogue. Precision and modalities are provider declarations, not independently tested. Published token quotes only.
0 IDs excluded by the documented scope checks. Rates below are alphabetical samples, per million tokens.
| Provider ID / delivery | Input /M | Output /M |
|---|---|---|
deepseek-ai/DeepSeek-V3.2-TEEbase · fp8 | USD 1.00 | USD 1.00 |
deepseek-ai/DeepSeek-V4-Flash-0731-TEEbase · fp8 | USD 0.44 | USD 1.32 |
google/gemma-4-31B-turbo-TEEbase · fp4 | USD 0.12 | USD 0.37 |
moonshotai/Kimi-K2.6-TEEbase · int4 | USD 0.50 | USD 2.85 |
Scaleway ↗
Observed 2026-10-07 · 13 provider IDs · 26 quotes
Source-currency research only. Chat models with both input and output rates. Region and standard/batch delivery remain separate. No USD conversion or leaderboard comparison.
3 IDs excluded by the documented scope checks. Rates below are alphabetical samples, per million tokens.
| Provider ID / delivery | Input /M | Output /M |
|---|---|---|
deepseek-v4-flash-0731batch · fr-par · precision unlisted | EUR 0.20 | EUR 0.40 |
deepseek-v4-flash-0731standard · fr-par · precision unlisted | EUR 0.40 | EUR 0.80 |
gemma-4-26b-a4b-itbatch · fr-par · precision unlisted | EUR 0.125 | EUR 0.25 |
gemma-4-26b-a4b-itstandard · fr-par · precision unlisted | EUR 0.25 | EUR 0.50 |
What the marketplace snapshot can support
789 of 1302 retained endpoint rows have no recognised precision label. Unknown precision cannot establish an equivalent substitute.
- Standard models with retained offers
- 348 / 358
- Distinct seller owners in those offers
- 75
- Covered models with competing sellers
- 142 / 348
- Covered models with two sellers at the same declared precision
- 69 / 348
Accepted offers dated 2026-10-06. The denominator is our published standard-model catalogue. A shared precision label still leaves context, modalities and service terms to check. This is coverage research, not a market census or a savings claim.
Download model-by-model coverage · JSONInspect definitions and replay instructions
| Seller | Models | Max spread |
|---|---|---|
| Relace | 9 | 488.89 |
| Reka | 1 | 41.42 |
| GMICloud | 5 | 14.37 |
| DekaLLM | 3 | 11.67 |
| DeepInfra | 22 | 11 |
| Darkbloom | 7 | 5 |
| Parasail | 4 | 5 |
| 11 | 4.43 | |
| OpenAI | 24 | 4.4 |
| Novita | 10 | 4.18 |
| CoreWeave | 1 | 3.26 |
| StreamLake | 4 | 2.7 |
| Alibaba | 7 | 2.45 |
| Inceptron | 2 | 2.34 |
| NextBit | 1 | 2.28 |
| DigitalOcean | 1 | 1.87 |
| SiliconFlow | 6 | 1.54 |
| Chutes | 1 | 1.5 |
| Venice | 3 | 1.4 |
| Minimax | 2 | 1.38 |
| Mancer 2 | 1 | 1.29 |
| Crusoe | 1 | 1.2 |
| xAI | 7 | 1.1 |
| Mistral | 14 | 1 |
| Azure | 13 | 1 |
| Google AI Studio | 9 | 1 |
| Claude Platform on AWS | 9 | 1 |
| Amazon Bedrock | 6 | 1 |
| Upstage | 2 | 1 |
| Tencent | 1 | 1 |
By vendor
Every publishable model from each vendor. Links go to price, scores, and dominance verdicts.
Aion Labs
6 entries · 0 with Arena scores · 6 with both token rates. Includes delivery variants.
Amazon
5 entries · 1 with Arena scores · 5 with both token rates. Includes delivery variants.
Anthracite Org
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Anthropic
29 entries · 15 with Arena scores · 29 with both token rates. Includes delivery variants.
- Claude Fable 5
- Claude Fable 5 (batch)(batch)
- Claude Fable 5.1
- Claude Fable 5.1 (batch)(batch)
- Claude Haiku 4.5
- Claude Haiku 4.5 (batch)(batch)
- Claude Opus 4.1
- Claude Opus 4.1 (batch)(batch)
- Claude Opus 4.5
- Claude Opus 4.5 (batch)(batch)
- Claude Opus 4.6
- Claude Opus 4.6 (batch)(batch)
- Claude Opus 4.7
- Claude Opus 4.7 (batch)(batch)
- Claude Opus 4.8
- Claude Opus 4.8 (batch)(batch)
- Claude Opus 5
- Claude Opus 5 (batch)(batch)
- Claude Opus 5.5
- Claude Opus 5.5 (batch)(batch)
- Claude Sonnet 4
- Claude Sonnet 4.5
- Claude Sonnet 4.5 (batch)(batch)
- Claude Sonnet 4.6
- Claude Sonnet 4.6 (batch)(batch)
- Claude Sonnet 5
- Claude Sonnet 5 (batch)(batch)
- Claude Sonnet 5.5
- Claude Sonnet 5.5 (batch)(batch)
Apodex
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
- Apodex 1.1 Mini (free)(free)
Arcee
1 entries · 1 with Arena scores · 1 with both token rates. Includes delivery variants.
Baidu
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
ByteDance
7 entries · 0 with Arena scores · 7 with both token rates. Includes delivery variants.
Cognitivecomputations
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Cohere
6 entries · 2 with Arena scores · 6 with both token rates. Includes delivery variants.
DeepSeek
15 entries · 11 with Arena scores · 15 with both token rates. Includes delivery variants.
Dots Studio
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Fireworks
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
42 entries · 15 with Arena scores · 40 with both token rates. Includes delivery variants.
- Gemini 2.5 Flash
- Gemini 2.5 Flash (batch)(batch)
- Gemini 2.5 Flash Lite
- Gemini 2.5 Flash Lite (batch)(batch)
- Gemini 2.5 Pro
- Gemini 2.5 Pro (batch)(batch)
- Gemini 2.5 Pro Preview 06-05
- Gemini 3 Flash Preview
- Gemini 3 Flash Preview (batch)(batch)
- Gemini 3.1 Flash Lite
- Gemini 3.1 Flash Lite (batch)(batch)
- Gemini 3.1 Flash Lite Preview
- Gemini 3.1 Pro Preview
- Gemini 3.1 Pro Preview (batch)(batch)
- Gemini 3.1 Pro Preview Custom Tools
- Gemini 3.5 Flash
- Gemini 3.5 Flash (batch)(batch)
- Gemini 3.5 Flash Lite
- Gemini 3.5 Flash Lite (batch)(batch)
- Gemini 3.6 Flash
- Gemini 3.6 Flash (batch)(batch)
- Gemini 3.7 Flash
- Gemini 3.7 Flash (batch)(batch)
- Gemini 3.8 Flash
- Gemini 3.8 Flash (batch)(batch)
- Gemma 2 27B
- Gemma 3 12B
- Gemma 3 27B
- Gemma 3 4B
- Gemma 4 26B A4B
- Gemma 4 26B A4B (free)(free)
- Gemma 4 31B
- Gemma 4 31B (free)(free)
- Lyria 3 Clip Preview
- Lyria 3 Pro Preview
- Nano Banana (Gemini 2.5 Flash Image)
- Nano Banana 2 (Gemini 3.1 Flash Image Preview)
- Nano Banana 2 (Gemini 3.1 Flash Image)
- Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)
- Nano Banana 2.1
- Nano Banana Pro (Gemini 3 Pro Image Preview)
- Nano Banana Pro (Gemini 3 Pro Image)
Gryphe
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
IBM
2 entries · 1 with Arena scores · 2 with both token rates. Includes delivery variants.
Inception
2 entries · 1 with Arena scores · 2 with both token rates. Includes delivery variants.
inclusionAI
5 entries · 0 with Arena scores · 5 with both token rates. Includes delivery variants.
Inference Net
2 entries · 0 with Arena scores · 2 with both token rates. Includes delivery variants.
Kwaipilot
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Liquid AI
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
- LFM2.5-2.6B (free)(free)
Mancer
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Meituan
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Meta
14 entries · 8 with Arena scores · 14 with both token rates. Includes delivery variants.
Microsoft
2 entries · 1 with Arena scores · 2 with both token rates. Includes delivery variants.
MiniMax
8 entries · 5 with Arena scores · 8 with both token rates. Includes delivery variants.
Mistral
25 entries · 3 with Arena scores · 25 with both token rates. Includes delivery variants.
- Codestral 2508
- Codestral 2508 (batch)(batch)
- Ministral 3 14B 2512
- Ministral 3 3B 2512
- Ministral 3 8B 2512
- Ministral 3 8B 2512 (batch)(batch)
- Mistral Large
- Mistral Large 2407
- Mistral Large 3 2512
- Mistral Large 3 2512 (batch)(batch)
- Mistral Large 4
- Mistral Medium 3
- Mistral Medium 3.1
- Mistral Medium 3.1 (batch)(batch)
- Mistral Medium 3.5
- Mistral Medium 3.5 (batch)(batch)
- Mistral Nemo
- Mistral Small 3
- Mistral Small 3.1 24B
- Mistral Small 3.2 24B
- Mistral Small 4
- Mistral Small 4 (batch)(batch)
- Mixtral 8x22B Instruct
- Saba
- Voxtral Small 24B 2507
Moonshot AI
8 entries · 3 with Arena scores · 8 with both token rates. Includes delivery variants.
Morph
2 entries · 0 with Arena scores · 2 with both token rates. Includes delivery variants.
Nex Agi
2 entries · 0 with Arena scores · 2 with both token rates. Includes delivery variants.
Nous Research
3 entries · 0 with Arena scores · 3 with both token rates. Includes delivery variants.
NVIDIA
11 entries · 0 with Arena scores · 10 with both token rates. Includes delivery variants.
OpenAI
102 entries · 30 with Arena scores · 102 with both token rates. Includes delivery variants.
- GPT Audio
- GPT Audio Mini
- GPT Chat Latest
- GPT-3.5 Turbo
- GPT-3.5 Turbo (batch)(batch)
- GPT-3.5 Turbo (older v0613)
- GPT-3.5 Turbo 16k
- GPT-3.5 Turbo Instruct
- GPT-4
- GPT-4 Turbo
- GPT-4 Turbo (batch)(batch)
- GPT-4.1
- GPT-4.1 (batch)(batch)
- GPT-4.1 Mini
- GPT-4.1 Mini (batch)(batch)
- GPT-4.1 Nano
- GPT-4.1 Nano (batch)(batch)
- GPT-4o
- GPT-4o (2024-05-13)
- GPT-4o (2024-08-06)
- GPT-4o (2024-11-20)
- GPT-4o (batch)(batch)
- GPT-4o-mini
- GPT-4o-mini (2024-07-18)
- GPT-4o-mini (batch)(batch)
- GPT-5
- GPT-5 (batch)(batch)
- GPT-5 Image
- GPT-5 Image Mini
- GPT-5 Mini
- GPT-5 Mini (batch)(batch)
- GPT-5 Nano
- GPT-5 Nano (batch)(batch)
- GPT-5 Pro
- GPT-5 Pro (batch)(batch)
- GPT-5.1
- GPT-5.1 (batch)(batch)
- GPT-5.1-Codex
- GPT-5.1-Codex-Max
- GPT-5.1-Codex-Mini
- GPT-5.2
- GPT-5.2 (batch)(batch)
- GPT-5.2 Chat
- GPT-5.2 Pro
- GPT-5.2 Pro (batch)(batch)
- GPT-5.2-Codex
- GPT-5.3-Codex
- GPT-5.4
- GPT-5.4 (batch)(batch)
- GPT-5.4 Image 2
- GPT-5.4 Mini
- GPT-5.4 Mini (batch)(batch)
- GPT-5.4 Nano
- GPT-5.4 Nano (batch)(batch)
- GPT-5.4 Pro
- GPT-5.4 Pro (batch)(batch)
- GPT-5.5
- GPT-5.5 (batch)(batch)
- GPT-5.5 Pro
- GPT-5.5 Pro (batch)(batch)
- GPT-5.6 Luna
- GPT-5.6 Luna (batch)(batch)
- GPT-5.6 Luna Pro
- GPT-5.6 Luna Pro (batch)(batch)
- GPT-5.6 Sol
- GPT-5.6 Sol (batch)(batch)
- GPT-5.6 Sol Pro
- GPT-5.6 Sol Pro (batch)(batch)
- GPT-5.6 Terra
- GPT-5.6 Terra (batch)(batch)
- GPT-5.6 Terra Pro
- GPT-5.6 Terra Pro (batch)(batch)
- GPT-6 Astra
- GPT-6 Astra (batch)(batch)
- GPT-6 Astra Pro
- GPT-6 Astra Pro (batch)(batch)
- GPT-6 Luna
- GPT-6 Luna (batch)(batch)
- GPT-6 Luna Pro
- GPT-6 Luna Pro (batch)(batch)
- GPT-6 Sol
- GPT-6 Sol (batch)(batch)
- GPT-6 Sol Pro
- GPT-6 Sol Pro (batch)(batch)
- GPT-6.1 Sol
- GPT-6.1 Sol Pro
- gpt-oss-120b
- gpt-oss-120b (batch)(batch)
- gpt-oss-20b
- gpt-oss-20b (batch)(batch)
- gpt-oss-safeguard-20b
- o1
- o1-pro
- o3
- o3 (batch)(batch)
- o3 Mini
- o3 Mini (batch)(batch)
- o3 Mini High
- o3 Pro
- o4 Mini
- o4 Mini (batch)(batch)
- o4 Mini High
OpenRouter
6 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Perceptron
2 entries · 0 with Arena scores · 2 with both token rates. Includes delivery variants.
Perplexity
5 entries · 0 with Arena scores · 5 with both token rates. Includes delivery variants.
Poolside
4 entries · 0 with Arena scores · 4 with both token rates. Includes delivery variants.
Prism Ml
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Qwen
53 entries · 22 with Arena scores · 53 with both token rates. Includes delivery variants.
- Qwen Plus 0728
- Qwen-Plus
- Qwen2.5 72B Instruct
- Qwen2.5 7B Instruct
- Qwen2.5 Coder 32B Instruct
- Qwen2.5 VL 72B Instruct
- Qwen3 14B
- Qwen3 235B A22B
- Qwen3 235B A22B Instruct 2507
- Qwen3 235B A22B Thinking 2507
- Qwen3 30B A3B
- Qwen3 30B A3B Instruct 2507
- Qwen3 30B A3B Thinking 2507
- Qwen3 32B
- Qwen3 8B
- Qwen3 Coder 30B A3B Instruct
- Qwen3 Coder 480B A35B
- Qwen3 Coder Flash
- Qwen3 Coder Next
- Qwen3 Coder Plus
- Qwen3 Max
- Qwen3 Max Thinking
- Qwen3 Next 80B A3B Instruct
- Qwen3 Next 80B A3B Thinking
- Qwen3 VL 235B A22B Instruct
- Qwen3 VL 235B A22B Thinking
- Qwen3 VL 30B A3B Instruct
- Qwen3 VL 30B A3B Thinking
- Qwen3 VL 32B Instruct
- Qwen3 VL 8B Instruct
- Qwen3 VL 8B Thinking
- Qwen3.5 397B A17B
- Qwen3.5 Plus 2026-02-15
- Qwen3.5 Plus 2026-04-20
- Qwen3.5-122B-A10B
- Qwen3.5-27B
- Qwen3.5-35B-A3B
- Qwen3.5-9B
- Qwen3.5-Flash
- Qwen3.6 27B
- Qwen3.6 35B A3B
- Qwen3.6 Flash
- Qwen3.6 Max Preview
- Qwen3.6 Plus
- Qwen3.7 Flash
- Qwen3.7 Max
- Qwen3.7 Plus
- Qwen3.8 2.4T A95B
- Qwen3.8 27B
- Qwen3.8 Flash
- Qwen3.8 Max (0902)
- Qwen3.8 Max Prime
- Qwen3.8 Omni Flash
Reka
2 entries · 0 with Arena scores · 2 with both token rates. Includes delivery variants.
Relace
2 entries · 0 with Arena scores · 2 with both token rates. Includes delivery variants.
Sakana AI
4 entries · 0 with Arena scores · 4 with both token rates. Includes delivery variants.
Sao10k
3 entries · 0 with Arena scores · 3 with both token rates. Includes delivery variants.
StepFun
2 entries · 1 with Arena scores · 2 with both token rates. Includes delivery variants.
Tencent
7 entries · 1 with Arena scores · 7 with both token rates. Includes delivery variants.
Thedrummer
3 entries · 0 with Arena scores · 3 with both token rates. Includes delivery variants.
Thinking Machines
4 entries · 2 with Arena scores · 4 with both token rates. Includes delivery variants.
Typesafe
1 entries · 0 with Arena scores · 0 with both token rates. Includes delivery variants.
Unbiased
2 entries · 0 with Arena scores · 2 with both token rates. Includes delivery variants.
Undi95
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
Upstage
3 entries · 1 with Arena scores · 3 with both token rates. Includes delivery variants.
Writer
1 entries · 0 with Arena scores · 1 with both token rates. Includes delivery variants.
xAI
8 entries · 4 with Arena scores · 8 with both token rates. Includes delivery variants.
Xiaomi
5 entries · 4 with Arena scores · 5 with both token rates. Includes delivery variants.
Z.ai
18 entries · 13 with Arena scores · 18 with both token rates. Includes delivery variants.