Celeris-1

By Celeris · released 2026-07-24 (6 weeks) · snapshot 2026-09-07

Off the frontierNot on frontier at launch dayproprietary
Intelligence index
7.60
Cost to run the AA index
$171
Price per 1M tokens (in / out)
$0.20 / $0.70
End-to-end time
0.95s
Better than…

Is Celeris-1 on the Pareto frontier, and what beats it?

Off the frontier

No. GPT-5.6 Luna (xhigh) is at least as smart for the same money or less.

Next step up: GPT-5.6 Luna (max), smarter but dearer.

Celeris-1 vs GPT-5.6 Luna (xhigh)Celeris-1 vs GPT-5.6 Luna (max)
#ModelCreatorIntelligenceTotal costE2E totalOpenReleasedFrontier
1Celeris-1Celeris7.60$1710.95s2026-07-24
2GPT-5.6 Luna (xhigh)OpenAI41.60$11966.5s2026-07-09Frontier
3GPT-5.6 Luna (max)OpenAI43.40$214139.8s2026-07-09Frontier

Open in the picker Daily check URL: 200 on the frontier, 409 once beaten

At a glance

Celeris-1 today

Time to first answer0.58s
Output speed1366 tok/s
Input modalitiesNot reported
0102030405060$10$20$50$100$200$500$1k$2k$5k$10kCost to run the Artificial Analysis Intelligence Index (USD, log scale)Intelligence IndexClaude 4.5 Haiku (Reasoning) (Anthropic): intelligence 22.50, $496Claude 4.5 Sonnet (Reasoning) (Anthropic): intelligence 27.30, $1123Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) (Anthropic): intelligence 53.20, $7603Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback) (Anthropic): intelligence 50.90, $1884Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback) (Anthropic): intelligence 52.80, $2534Claude Opus 4.8 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 47.80, $4565Claude Opus 5 (Adaptive Reasoning, High Effort) (Anthropic): intelligence 52.00, $3016Claude Opus 5 (Adaptive Reasoning, Low Effort) (Anthropic): intelligence 43.80, $811Claude Opus 5 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 54.10, $5546Claude Opus 5 (Adaptive Reasoning, Medium Effort) (Anthropic): intelligence 49.50, $1688Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) (Anthropic): intelligence 53.40, $4299Claude Sonnet 5 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 45.10, $5313DeepSeek V3 (Dec '24) (DeepSeek): intelligence 10.40, $26.2DeepSeek V4 Flash 0731 (Reasoning, Max Effort) (DeepSeek): intelligence 40.80, $363DeepSeek V4 Flash Vision (Reasoning, Max Effort) (DeepSeek): intelligence 40.70, $308DeepSeek V4 Pro (Reasoning, Max Effort) (DeepSeek): intelligence 34.50, $184DeepSeek V4 Pro 0813 (Reasoning, Max Effort) (DeepSeek): intelligence 42.10, $707Gemini 3.1 Flash-Lite (Google): intelligence 19.30, $96.2Gemini 3.1 Pro Preview (Google): intelligence 36.70, $880Gemini 3.5 Flash (high) (Google): intelligence 39.70, $1420Gemini 3.5 Flash-Lite (Google): intelligence 27.60, $166Gemini 3.6 Flash (high) (Google): intelligence 40.30, $509Gemini 3.7 Flash (high) (Google): intelligence 45.20, $688Gemini 3.8 Flash (high) (Google): intelligence 47.10, $1078GLM-5.2 (max) (Z AI): intelligence 42.10, $1006GLM-5.3 (max) (Z AI): intelligence 48.60, $1762GPT-5.5 (medium) (OpenAI): intelligence 41.20, $1201GPT-5.6 Sol (high) (OpenAI): intelligence 48.30, $953GPT-5.6 Sol (low) (OpenAI): intelligence 40.80, $355GPT-5.6 Sol (max) (OpenAI): intelligence 51.30, $2464GPT-5.6 Sol (medium) (OpenAI): intelligence 46.00, $580GPT-5.6 Sol (xhigh) (OpenAI): intelligence 49.80, $1459GPT-5.6 Terra (high) (OpenAI): intelligence 41.30, $516GPT-5.6 Terra (max) (OpenAI): intelligence 46.80, $1714GPT-5.6 Terra (xhigh) (OpenAI): intelligence 44.40, $806GPT-6 Astra (medium) (OpenAI): intelligence 52.20, $1559GPT-6 Astra (Non-reasoning) (OpenAI): intelligence 47.80, $1631gpt-oss-120b (high) (OpenAI): intelligence 15.60, $146gpt-oss-20b (high) (OpenAI): intelligence 10.80, $33.7Granite 4.2 8B (IBM): intelligence 13.90, $40.9Grok 4.3 (high) (SpaceXAI): intelligence 29.90, $310Grok 4.5 (high) (SpaceXAI): intelligence 45.50, $759Grok 4.6 (high) (SpaceXAI): intelligence 50.60, $1632Grok 4.6 (low) (SpaceXAI): intelligence 41.60, $343Grok 4.6 (medium) (SpaceXAI): intelligence 48.80, $1262Grok 4.6 (xhigh) (SpaceXAI): intelligence 49.90, $2026Inkling (xhigh) (Thinking Machines): intelligence 32.20, $826Inkling Small (Thinking Machines): intelligence 32.20, $203Kimi K2.7 Code (Kimi): intelligence 32.70, $623Kimi K3 (low) (Kimi): intelligence 37.10, $379Kimi K3 (max) (Kimi): intelligence 50.20, $3070Ling 3.0 Flash (InclusionAI): intelligence 27.40, $75.8Llama 4 Maverick (Meta): intelligence 10.30, $51.1MiniMax-M3 (MiniMax): intelligence 35.70, $293Mistral Large 3 (Mistral): intelligence 11.10, $91.2Mistral Small 3.1 (Mistral): intelligence 10.30, $38.1Muse Glimmer (high) (Meta): intelligence 24.40, $101Muse Spark 1.1 (xhigh) (Meta): intelligence 41.20, $789Nemotron 3 Ultra 550B A55B (Reasoning) (NVIDIA): intelligence 29.30, $584Quasar 438B (max, based on GLM-5.2) (Multiverse Computing): intelligence 34.20, $1377Qwen3 Coder Next (Alibaba): intelligence 14.10, $388Qwen3.7 Max (Alibaba): intelligence 36.60, $1280Qwen3.7 Plus (Alibaba): intelligence 31.90, $310Qwen3.8 2.4T A95B (Alibaba): intelligence 46.70, $1780Qwen3.8 27B (xhigh) (Alibaba): intelligence 41.40, $900Qwen3.8 Max (Alibaba): intelligence 46.90, $1938Solar Pro 3 (Upstage): intelligence 9.40, $155Trinity Large Thinking (Arcee AI): intelligence 13.10, $327Granite 4.2 3B (IBM): intelligence 10.70, $13.6MiMo-V2.5 (Xiaomi): intelligence 28.20, $26.4DeepSeek V4 Flash (Reasoning, High Effort) (DeepSeek): intelligence 30.20, $96.8DeepSeek V4 Flash (Reasoning, Max Effort) (DeepSeek): intelligence 30.70, $101Hy3 (Tencent): intelligence 32.40, $102MiMo-V2.5-Pro (Xiaomi): intelligence 32.60, $105GPT-5.6 Luna (xhigh) (OpenAI): intelligence 41.60, $119GPT-5.6 Luna (max) (OpenAI): intelligence 43.40, $214GLM-5.3-Flash (Z AI): intelligence 46.20, $219Muse Spark 1.2 (xhigh) (Meta): intelligence 46.80, $822GPT-6 Astra (low) (OpenAI): intelligence 49.30, $873Muse Spark 1.3 (xhigh) (Meta): intelligence 51.60, $1129Muse Spark 1.3 (max) (Meta): intelligence 53.00, $1341GPT-6 Astra (high) (OpenAI): intelligence 53.40, $2039GPT-6 Astra (xhigh) (OpenAI): intelligence 54.30, $2778Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback) (Anthropic): intelligence 54.50, $3603GPT-6 Astra (max) (OpenAI): intelligence 54.70, $4064Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) (Anthropic): intelligence 56.20, $7037Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) (Anthropic): intelligence 56.80, $10816Celeris-1 (Celeris): intelligence 7.60, $171Celeris-1Granite 4.2 3BGPT-5.6 Luna (xhigh)GPT-5.6 Luna (max)Claude Fable 5.1 (max)
Celeris-1Same release, other effort levelsToday's Pareto frontierEvery other model
Up and left is better. Dotted line: today's Pareto frontier. Hover a point for its numbers, click it for its page.

Intelligence: Artificial Analysis Intelligence Index. Cost: total cost to run that benchmark, so reasoning tokens count. API id: celeris-1.

Matchup

Compare head to head

Suggested: what beats it, its effort levels, and frontier models near its price.

Compare Celeris-1 with measured models

Release

What the Celeris-1 release did to the frontier on 2026-07-24

None of its 1 effort levels reached the frontier at launch.

Closest miss: Celeris-1, 2.8 points below DeepSeek V3 (Dec '24) at 6.5× the cost.

Release page: launch-day frontier and every effort level

Deciding

Should you use Celeris-1?

Should you use Celeris-1?

Only when it fits your setup: GPT-5.6 Luna (xhigh) is at least as smart for the same money or less, so Celeris-1 is off the value frontier today.

What is better and cheaper than Celeris-1?

GPT-5.6 Luna (xhigh) — at least as smart for the same money or less.

How much does Celeris-1 cost?

$0.20/$0.70 in/out per 1M tokens, and $171 to run the full Artificial Analysis Intelligence Index, the honest all-in number for reasoning models.

Is Celeris-1 on the Pareto frontier?

No: GPT-5.6 Luna (xhigh) beats it on intelligence, cost, or both.

Frontier membership by axis pair

Other axis pairs where Celeris-1 is Pareto-optimal.

Y axisX axisOn the frontier
AA intelligence indexAA total cost to run (USD)
AA intelligence indexEnd-to-end total seconds
AA intelligence indexOutput speed tokens/sec
Agents: Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/models/celeris-1 · Markdown: /api/v1/models/celeris-1?format=md · llms.txt · agent skill