Compute
Compute is the only currency inside Nexara. One unit of Compute is worth $0.0000008 — so 100,000 Compute = $0.08 and 1,000,000 Compute = $0.80. You subscribe to a plan for a Daily Compute limit or earn free Compute — and every model you talk to is billed from that same pool at the model's own rate.
1M Compute = $0.80
The fixed exchange rate — the dollar figure is only ever an internal basis; you always see Compute.
Per-model rates
Every model bills its actual provider rate — a cheap flash model costs a fraction of a frontier flagship.
12,500 Compute per image
AI image generation (Qwen 3.8 Max) costs a flat 12,500 Compute per image, regardless of size.
MiniMax M3 Compute pricing
MiniMax M3 is a paid model and is billed from Compute at its listed input and output rates.
How the allowance works
- Daily Compute limit. Your shared daily allowance resets at 00:00 UTC. Every model draws from the same allowance at its own rate.
- Weekly Compute limit. A shared weekly allowance is worth 1.5× your plan price (2× on web chat) and resets every Monday.
- A hard stop. There is no separate usage balance. When the Daily Compute limit or Weekly Compute limit is reached, Nexara pauses new requests until the relevant reset.
- Cached tokens are cheaper. When a model re-reads part of your conversation from its prompt cache, those input tokens bill at the model's cache-read rate — often a fraction of the full input rate.
Model pricing
All 120 models in the catalog, with their Compute cost per million tokens — input, output, and cached input. OpenRouter Free Route, AutoRouter, Ling 3.0 Flash Finance, Dots 3 Note (Preview), North Mini Code, Nemotron Rerank VL 1B v2, Nemotron 3.5 Content Safety, Nemotron Embed VL 1B v2, Nemotron 3 Embed 1B have a zero Compute input rate. They are still part of Nexara's Compute accounting, and any output or other listed rate can still apply. Models marked Locked are temporarily unavailable at the provider — they stay listed for reference but can't be selected until it recovers.
Frontier models: Grok 4.5, Grok 4.6, GPT-5.6 Terra, GPT-5.6 Sol, GPT-6 Astra, Claude Sonnet 5, Claude Opus 4.8, Claude Opus 5, Claude Fable 5, Fable 5.1, Qwen 3.8 Omni Flash, Qwen 3.8 Max, Qwen 3.7 Max, Qwen 3.6 Max (Preview), GLM 5.1, GLM 5.2, GLM 5.3 are marked Frontier and use the same daily and weekly Compute allowance as every other model. Their provider pricing determines how quickly that shared allowance is used. Models marked Advanced are strong, newer models that stay on regular Compute like any other model; the label is just a quality signal.
About the model IDs: catalog IDs follow the router convention provider/name (e.g. anthropic/claude-sonnet-4.6, openai/gpt-5.6-luna) — the same format routers like OpenRouter use. The prefix identifies the provider family a model belongs to, not a claim that the provider publishes that exact retail name. Models are served through a mix of direct provider connections (Claude models through Anthropic, GPT models through OpenAI) and open model routers, depending on the model, and the chat always shows you which model actually served your reply.
Anthropic
9 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Fable 5.1FrontierFrontier anthropic/claude-fable-5-1 | 12.5M | 62.5M | 312.5K | 1M | ||
02Claude Opus 5FrontierFrontier anthropic/claude-opus-5 | 6.25M | 31.25M | 625K | 1M | ||
03Claude Fable 5FrontierFrontier anthropic/claude-fable-5 | 12.5M | 62.5M | 1.25M | 1M | ||
04Claude Opus 4.8FrontierFrontier anthropic/claude-opus-4.8 | 6.25M | 31.25M | 625K | 1M | ||
05Claude Opus 4.7AdvancedFrontier anthropic/claude-opus-4.7 | 6.25M | 31.25M | 625K | 1M | ||
06Claude Sonnet 5FrontierFrontier anthropic/claude-sonnet-5 | 2.5M | 12.5M | 250K | 1M | ||
07Claude Sonnet 4.6AdvancedFrontier anthropic/claude-sonnet-4.6 | 3.75M | 18.75M | 375K | 1M | ||
08Claude Opus 4.6AdvancedFrontier anthropic/claude-opus-4.6 | 6.25M | 31.25M | 625K | 1M | ||
09Claude Haiku 4.5Speed anthropic/claude-haiku-4.5 | 1.25M | 6.25M | 125K | 200K |
Atria
1 model| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Atria Dawn PreviewGeneral atria/Atria-Dawn-Preview | 125K | 125K | — | 256K |
CCohere
1 model| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01North Mini CodeFreeCoding cohere/north-mini-code:free | Free | Free | — | 262K |
DeepSeek
6 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01DeepSeek V4 ProReasoning deepseek/deepseek-v4-pro | 750K | 3M | — | 1M | ||
02DeepSeek V4 FlashAdvancedReasoning deepseek/deepseek-v4-flash | 312.5K | 1.25M | — | 1M | ||
03DeepSeek V4 Flash 07.31Reasoning deepseek/deepseek-v4-flash-0731 | 50K | 162.5K | 11.3K | 1M | ||
04DeepSeek V4.1 FlashSpeed deepseek/deepseek-v4.1-flash:free | 125K | 487.5K | 125K | 64K | ||
05DeepSeek V3.2Reasoning deepseek/deepseek-v3.2 | 350K | 525K | — | 131K | ||
06DeepSeek Chat V3.1General deepseek/deepseek-chat-v3.1 | 350K | 525K | — | 164K |
DSDots Studio
1 model| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Dots 3 Note (Preview)FreeGeneral dots-studio/dots-3-note-preview:free | Free | Free | — | 512K |
GLM
10 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01GLM 5.3FrontierFrontier z-ai/glm-5.3 | 1.75M | 5.5M | 325K | 1M | ||
02GLM 5.2FrontierFrontier z-ai/glm-5.2 | 1.88M | 5.65M | — | 1M | ||
03GLM 5.1FrontierFrontier z-ai/glm-5.1 | 1.78M | 5.53M | — | 200K | ||
04GLM 5AdvancedFrontier z-ai/glm-5 | 787.5K | 2.44M | — | 200K | ||
05GLM 5 TurboAdvancedSpeed z-ai/glm-5-turbo | 1.55M | 5.05M | — | 200K | ||
06GLM 4.7AdvancedReasoning z-ai/glm-4.7 | 800K | 2.8M | — | 200K | ||
07GLM 4.6Coding z-ai/glm-4.6 | 775K | 2.78M | — | 200K | ||
08GLM 4.5General z-ai/glm-4.5 | 787.5K | 2.79M | — | 131K | ||
09GLM 4.5 AirSpeed z-ai/glm-4.5-air | 275K | 1.4M | — | 131K | ||
10GLM 5.3 FlashSpeed z-ai/glm-5.3-flash | 187.5K | 625K | 37.5K | 1M |
| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Gemini 3.7 FlashLockedMultimodal google/gemini-3.7-flash | 475K | 2.35M | 50K | 1M | ||
02Gemini 3.6 FlashLockedMultimodal google/gemini-3.6-flash | 1.88M | 9.38M | — | 1M | ||
03Gemini 3.5 FlashLockedMultimodal google/gemini-3.5-flash | 1.88M | 11.25M | — | 1M | ||
04Gemini 3.1 ProLockedMultimodal google/gemini-3.1-pro | 2.5M | 15M | — | 1M | ||
05Gemini 2.5 ProLockedMultimodal google/gemini-2.5-pro | 1.56M | 12.5M | — | 1M | ||
06Gemini 3 FlashLockedMultimodal google/gemini-3-flash | 625K | 3.75M | — | 1M | ||
07Gemini 2.5 FlashLockedMultimodal google/gemini-2.5-flash | 375K | 3.13M | — | 1M | ||
08Gemma 4 31BGeneral google/gemma-4-31b-it:free | 187.5K | 750K | — | 262K | ||
09Gemma 4 26B A4BGeneral google/gemma-4-26b-a4b-it:free | 150K | 500K | — | 262K |
Inception
2 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Mercury 2.5Reasoning inception/mercury-2.5 | 50K | 187.5K | 5K | 260K | ||
02Mercury 2Reasoning inception/mercury-2 | 312.5K | 937.5K | 31.3K | 128K |
Inclusion AI
2 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Ling 3.0 FlashGeneral inclusion-ai/ling-3.0-flash | 62.5K | 187.5K | — | 260K | ||
02Ling 3.0 Flash FinanceFreeGeneral inclusionai/ling-3.0-flash-fin:free | Free | Free | — | 262K |
LLiquid
1 model| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01LFM 2.5 2.6BSpeed liquid/lfm-2.5-2.6b:free | 12.5K | 25K | — | 131K |
Meta
4 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Llama 3.3 70B InstructGeneral meta/llama-3.3-70b-instruct | 187.5K | 750K | — | 131K | ||
02Llama 3.1 8B InstructGeneral meta/llama-3.1-8b-instruct | 50K | 125K | — | 16K | ||
03Llama 3.2 3B InstructSpeed meta/llama-3.2-3b-instruct | 37.5K | 100K | — | 16K | ||
04Llama 3.2 1B InstructSpeed meta/llama-3.2-1b-instruct | 25K | 62.5K | — | 16K |
MiniMax
8 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01MiniMax M3Frontier minimax/minimax-m3 | 375K | 1.5M | — | 1M | ||
02MiniMax M2.7AdvancedGeneral minimax/minimax-m2.7:free | 375K | 1.5M | — | 205K | ||
03MiniMax M2.7 High-SpeedSpeed minimax/minimax-m2.7-highspeed:free | 187.5K | 750K | — | 205K | ||
04MiniMax M2.5General minimax/minimax-m2.5:free | 375K | 1.5M | — | 205K | ||
05MiniMax M2.5 High-SpeedSpeed minimax/minimax-m2.5-highspeed:free | 187.5K | 750K | — | 205K | ||
06MiniMax M2.1General minimax/minimax-m2.1:free | 375K | 1.5M | — | 205K | ||
07MiniMax M2.1 High-SpeedSpeed minimax/minimax-m2.1-highspeed:free | 187.5K | 750K | — | 205K | ||
08MiniMax M2General minimax/minimax-m2:free | 375K | 1.5M | — | 205K |
Mistral
8 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Mistral Large 3General mistralai/mistral-large-2512 | 2.5M | 7.5M | — | 256K | ||
02Mistral Medium 3.5AdvancedGeneral mistralai/mistral-medium-3.5 | 437.5K | 1.31M | — | 256K | ||
03Devstral MediumCoding mistralai/devstral-medium | 625K | 812.5K | — | 256K | ||
04Codestral 25.08Coding mistralai/codestral-2508 | 375K | 1.13M | — | 256K | ||
05Mistral Small 26.03General mistralai/mistral-small-2603 | 125K | 375K | — | 256K | ||
06Ministral 14BGeneral mistralai/ministral-14b | 250K | 250K | — | 256K | ||
07Ministral 8BSpeed mistralai/ministral-8b | 125K | 125K | — | 256K | ||
08Ministral 3BSpeed mistralai/ministral-3b | 62.5K | 62.5K | — | 128K |
MMoonshot
4 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Kimi K3Frontier moonshotai/kimi-k3 | 3.75M | 18.75M | 375K | 262K | ||
02Kimi K2.7 CodeLockedCoding moonshotai/kimi-k2.7-code | 875K | 4.38M | 187.5K | 262K | ||
03Kimi K2.6LockedFrontier moonshotai/kimi-k2.6 | 1.19M | 5M | — | 262K | ||
04Kimi K2.5LockedReasoning moonshotai/kimi-k2.5 | 712.5K | 3.56M | — | 262K |
NVIDIA
15 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Nemotron 3 UltraReasoning nvidia/nemotron-3-ultra | 625K | 2.5M | — | 1M | ||
02Nemotron 3 SuperReasoning nvidia/nemotron-3-super | 312.5K | 1.25M | — | 1M | ||
03Llama 3.3 Nemotron Super 49BGeneral nvidia/llama-3.3-nemotron-super-49b | 187.5K | 750K | — | 131K | ||
04Nemotron 3 Nano OmniMultimodal nvidia/nemotron-3-nano-omni | 62.5K | 187.5K | — | 256K | ||
05Nemotron 3 Nano 30B A3BReasoning nvidia/nemotron-3-nano-30b-a3b | 62.5K | 187.5K | — | 256K | ||
06Nemotron Nano 9B V2Speed nvidia/nemotron-nano-9b-v2 | 37.5K | 125K | — | 128K | ||
07Nemotron 3 NanoReasoning nvidia/nemotron-3-nano | 125K | 500K | — | 1M | ||
08Nemotron 3 Ultra 550BFrontier nvidia/nemotron-3-ultra-550b-a55b:free | 1.25M | 3.75M | — | 1M | ||
09Nemotron 3.5 LightningSpeed nvidia/nemotron-3.5-lightning:free | 87.5K | 275K | — | 262K | ||
10Nemotron 3 Super 120BGeneral nvidia/nemotron-3-super-120b-a12b:free | 112.5K | 525K | — | 1M | ||
11Nemotron 3 Nano Omni ReasoningReasoning nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free | 100K | 375K | — | 262K | ||
12Nemotron Rerank VL 1B v2FreeGeneral nvidia/llama-nemotron-rerank-vl-1b-v2:free | Free | Free | — | 128K | ||
13Nemotron 3.5 Content SafetyFreeGeneral nvidia/nemotron-3.5-content-safety:free | Free | Free | — | 128K | ||
14Nemotron Embed VL 1B v2FreeGeneral nvidia/llama-nemotron-embed-vl-1b-v2:free | Free | Free | — | 131K | ||
15Nemotron 3 Embed 1BFreeGeneral nvidia/nemotron-3-embed-1b:free | Free | Free | — | 33K |
OpenAI
6 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01GPT-6 AstraFrontierFrontier openai/gpt-6-astra | 12.5M | 62.5M | 1.25M | 1M | ||
02GPT-5.6 SolFrontierFrontier openai/gpt-5.6-sol | 6.25M | 37.5M | 625K | 1M | ||
03GPT-5.6 TerraFrontierFrontier openai/gpt-5.6-terra | 2.56M | 14.69M | — | 1M | ||
04GPT-5.6 LunaAdvancedFrontier openai/gpt-5.6-luna | 250K | 1.31M | — | 1M | ||
05GPT-OSS-120BFrontier openai/gpt-oss-120b | 125K | 500K | — | 131K | ||
06GPT-5.3 Codex SparkCoding openai/gpt-5.3-codex-spark | 187.5K | 750K | — | 128K |
Poolside
3 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Laguna S 2.1Coding poolside/laguna-s-2.1:free | 112.5K | 225K | — | 1M | ||
02Laguna XS 2.1Coding poolside/laguna-xs-2.1:free | 75K | 150K | — | 262K | ||
03Laguna XS.2Coding poolside/laguna-xs.2 | 125K | 500K | — | 131K |
Qwen
18 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Qwen 3.8 MaxFrontierFrontier qwen/qwen3.8-max | 3.19M | 9.44M | — | 1M | ||
02Qwen 3.8 Omni FlashFrontierMultimodal qwen/qwen3.8-omni-flash | 187.5K | 587.5K | 20K | 1M | ||
03Qwen 3.7 MaxFrontierFrontier qwen/qwen3.7-max | 3.19M | 9.44M | — | 1M | ||
04Qwen 3.7 PlusAdvancedGeneral qwen/qwen3.7-plus | 562.5K | 2.06M | — | 1M | ||
05Qwen 3.6 Max (Preview)FrontierFrontier qwen/qwen3.6-max-preview | 1.69M | 9.81M | — | 256K | ||
06Qwen 3.6 PlusAdvancedGeneral qwen/qwen3.6-plus | 687.5K | 3.81M | — | 1M | ||
07Qwen 3.6 27BReasoning qwen/qwen3.6-27b | 812.5K | 4.56M | — | 256K | ||
08Qwen 3.5 397B A17BFrontier qwen/qwen3.5-397b-a17b | 812.5K | 4.56M | — | 256K | ||
09Qwen 3.5 PlusAdvancedGeneral qwen/qwen3.5-plus | 562.5K | 3.06M | — | 1M | ||
10Qwen 3.5 Omni PlusMultimodal qwen/qwen3.5-omni-plus | 1.81M | 13.81M | — | 128K | ||
11Qwen3 Coder PlusCoding qwen/qwen3-coder-plus | 1.31M | 6.31M | — | 1M | ||
12Qwen3 MaxFrontier qwen/qwen3-max | 1.56M | 7.56M | — | 256K | ||
13Qwen3 VL PlusMultimodal qwen/qwen3-vl-plus | 312.5K | 2.06M | — | 256K | ||
14Qwen 3.6 35B A3BReasoning qwen/qwen3.6-35b-a3b | 372.5K | 1.92M | — | 256K | ||
15Qwen 3.5 FlashSpeed qwen/qwen3.5-flash | 187.5K | 562.5K | — | 1M | ||
16Qwen 3.5 Omni FlashMultimodal qwen/qwen3.5-omni-flash | 562.5K | 3.81M | — | 128K | ||
17Qwen3 Omni FlashMultimodal qwen/qwen3-omni-flash | 600K | 2.14M | — | 128K | ||
18Qwen Plus 07.28General qwen/qwen-plus-2025-07-28 | 562.5K | 1.56M | — | 1M |
Router
2 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01AutoRouterFreeGeneral router/autorouter | Free | Free | — | 131K | ||
02OpenRouter Free RouteFreeGeneral router/openrouter-free | Free | Free | — | 131K |
SSenseNova
2 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01SenseNova 6.8 Flash-LiteGeneral sensenova/sensenova-6.8-flash-lite | 25K | 100K | — | 262K | ||
02SenseNova 6.7 Flash-LiteGeneral sensenova/sensenova-6.7-flash-lite | 25K | 100K | — | 262K |
StepFun
1 model| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Step 3.7 FlashReasoning stepfun/step-3.7-flash | 125K | 375K | — | 256K |
TMThinking Machines
2 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01InklingReasoning thinkingmachines/inkling:free | 1.19M | 5.06M | — | 1M | ||
02Inkling SmallReasoning thinkingmachines/inkling-small:free | 562.5K | 1.5M | — | 1M |
xAI
3 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Grok 4.6FrontierFrontier x-ai/grok-4.6 | 2.5M | 7.5M | — | 500K | ||
02Grok 4.5FrontierFrontier x-ai/grok-4.5 | 2.18M | 6.59M | — | 500K | ||
03Grok Build 0.1Coding x-ai/grok-build-0.1 | 1.25M | 2.5M | 250K | 256K |
Xiaomi
2 models| Model | Compute / 1M in | Compute / 1M out | Compute / 1M cache | Image | Video | Context |
|---|---|---|---|---|---|---|
01Xiaomi Mimo V2.5 ProLockedReasoning xiaomi/mimo-v2.5-pro:free | 537.5K | 1.09M | — | 128K | ||
02Xiaomi Mimo V2.5LockedReasoning xiaomi/mimo-v2.5:free | 537.5K | 1.09M | — | 128K |
Prices are Compute per million tokens at each model's actual provider rate. The cache column only appears when the model's provider advertises a separate cache-read rate — otherwise cached input bills at the full input rate. Context is the model's maximum window in tokens.
Image & video input
Some models accept media directly in the conversation — not just text. Image input means you can attach a picture and have the model read it (screenshots, charts, documents). Video input goes further: the model understands frames of a video clip. Media input is billed at the same token rate as text — vision tokens count toward your Compute like any other input.
Image input60
OpenRouter Free RouteMiniMax M3Ministral 3BMinistral 8BMinistral 14BMistral Small 26.03Mistral Medium 3.5Mistral Large 3Step 3.7 FlashGrok 4.5Grok 4.6Grok Build 0.1GPT-5.6 LunaGPT-5.6 TerraGPT-5.6 SolGPT-6 AstraClaude Haiku 4.5Claude Sonnet 4.6Claude Sonnet 5Claude Opus 4.6Claude Opus 4.7Claude Opus 4.8Claude Opus 5Claude Fable 5Fable 5.1Kimi K3Kimi K2.7 CodeKimi K2.6Kimi K2.5Gemini 3.6 FlashGemini 3.7 FlashGemini 3.5 FlashGemini 3.1 ProGemini 3 FlashGemini 2.5 FlashGemini 2.5 ProQwen 3.8 Omni FlashQwen 3.8 MaxQwen 3.7 PlusQwen 3.6 PlusQwen 3.6 27BQwen 3.6 35B A3BQwen 3.5 PlusQwen 3.5 397B A17BQwen 3.5 Omni PlusQwen 3.5 FlashQwen 3.5 Omni FlashQwen3 VL PlusQwen3 Omni FlashGLM 5.3 FlashDeepSeek V4.1 FlashNemotron 3 Nano OmniSenseNova 6.7 Flash-LiteSenseNova 6.8 Flash-LiteNemotron 3 Nano Omni ReasoningNemotron Rerank VL 1B v2Gemma 4 31BNemotron 3.5 Content SafetyNemotron Embed VL 1B v2Gemma 4 26B A4B
Video input11
Gemini 3.6 FlashGemini 3.7 FlashGemini 3.5 FlashGemini 3.1 ProGemini 3 FlashGemini 2.5 FlashGemini 2.5 ProQwen 3.8 Omni FlashQwen 3.5 Omni PlusQwen 3.5 Omni FlashQwen3 Omni Flash
Frequently asked
What is one unit of Compute worth?
$0.0000008 — so 100,000 Compute is $0.08 and 1,000,000 Compute is $0.80. All prices on this page are Compute per million tokens at that fixed rate.
Why do different models cost different amounts of Compute?
Each model bills at its actual provider rate. A tiny flash model costs a fraction of a frontier flagship because that's what the provider charges Nexara — the Compute price is that rate scaled to the 1M-Compute-per-dollar ratio.
Is any model free?
Some models may show a zero input rate, but “free” does not mean outside Nexara's accounting. Every model in the catalog is billed from Compute according to its listed input, output, and cached-input rates; check the table above for the current rate of a specific model.
What is the cache rate?
When a conversation repeats the same system prompt and earlier messages, the provider can serve that prefix from its prompt cache. Those cached input tokens bill at the model's cache-read rate (shown in the table when a model has one) — usually 5–10× cheaper than fresh input.
What happens when I run out of Compute?
You're blocked from new requests until the Daily Compute limit refills or the Weekly Compute limit resets Monday. Settings shows usage percentages and reset timing, and Nexara refuses a request before it starts when the allowance is exhausted.
For developers
Building something with Nexara or NexaraClaw? The Agent Visualization SDK connects your own front-end — a game-like NPC view, a voice assistant UI — to your local agents over a small WebSocket protocol.