Claude Opus 4.6
Opus 4.6 is the world’s best model for coding and professional work, built to power agents that take on whole categories of real-world work.
text + image + file → text · made by Anthropic
$5→$25per Mtok in / out
Anthropic · 13 sellers
Sold by 26 ways
| Seller | Lane | Rate |
|---|---|---|
| AWS Bedrock 0 ways | $2.5per Mtok in$12.5per Mtok out | |
| AWS Bedrock cloudnot on the seller's list today | batch | $2.5per Mtok in$12.5per Mtok outaws.amazon.com · read 2026-08-24 |
| AWS Bedrock 0 ways | $2.5per Mtok in$12.5per Mtok out | |
| AWS Bedrock cloudnot on the seller's list today | batch cross-region | $2.5per Mtok in$12.5per Mtok outaws.amazon.com · read 2026-08-24 |
| Anthropic api | batch | $2.5per Mtok in$12.5per Mtok outplatform.claude.com · read 2026-08-24 |
| Google Vertex AI 0 ways | $2.5per Mtok in$12.5per Mtok out$0.25per Mtok cached$3.12per Mtok cache write | |
| Google Vertex AI cloudnot on the seller's list today | batch | $2.5per Mtok in$12.5per Mtok out$0.25per Mtok cached$3.12per Mtok cache writecloud.google.com · read 2026-08-24 |
| AWS Bedrock cloud | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeaws.amazon.com · read 2026-08-24 |
| AWS Bedrock 0 ways | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache write | |
| AWS Bedrock cloudnot on the seller's list today | cross-region | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeaws.amazon.com · read 2026-08-24 |
| Anthropic api | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeplatform.claude.com · read 2026-08-24 |
| Databricks cloud | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writemodels.dev · read 2026-09-16 |
| DigitalOcean cloud | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writemodels.dev · read 2026-09-16 |
| Google Vertex AI cloud | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writecloud.google.com · read 2026-08-24 |
| Microsoft Azure AI cloud | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writemodels.dev · read 2026-09-16 |
| Nous Research 2 ways | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache write | |
| Nous Research aggregator | batch | $2.5per Mtok in$12.5per Mtok out$0.25per Mtok cached$3.12per Mtok cache writeinference-api.nousresearch.com · read 2026-09-12 |
| Nous Research aggregator | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeinference-api.nousresearch.com · read 2026-09-12 |
| OpenRouter 3 ways | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache write | |
| OpenRouter aggregator | batch | $2.5per Mtok in$12.5per Mtok out$0.25per Mtok cached$3.12per Mtok cache writeopenrouter.ai · read 2026-08-24 |
| OpenRouter aggregator | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeopenrouter.ai · read 2026-08-24 |
| OpenRouter aggregator | google-vertex/europe | $5.5per Mtok in$27.5per Mtok out$0.55per Mtok cached$6.88per Mtok cache writeopenrouter.ai · read 2026-08-25 |
| Requesty 3 ways | $4.95per Mtok in$24.75per Mtok out$0.49per Mtok cached | |
| Requesty aggregatornot on the seller's list today | vertex europe-west1 | $4.95per Mtok in$24.75per Mtok out$0.49per Mtok cachedrouter.requesty.ai · read 2026-08-25 |
| Requesty aggregatornot on the seller's list today | vertex us-east5 | $4.95per Mtok in$24.75per Mtok out$0.49per Mtok cachedrouter.requesty.ai · read 2026-08-25 |
| Requesty aggregator | anthropic | $5per Mtok in$25per Mtok out$0.5per Mtok cachedrouter.requesty.ai · read 2026-08-28 |
| Requesty aggregator | bedrock | $5per Mtok in$25per Mtok out$0.5per Mtok cachedrouter.requesty.ai · read 2026-08-28 |
| Requesty aggregator | vertex | $5.5per Mtok in$27.5per Mtok out$0.55per Mtok cachedrouter.requesty.ai · read 2026-09-16 |
| ElectronHub aggregator | standard | $5per Mtok in$25per Mtok outapi.electronhub.ai · read 2026-09-16 |
| GitHub Copilot 0 ways | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache write | |
| GitHub Copilot aggregatornot on the seller's list today | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writemodels.dev · read 2026-09-04 |
| Inference.net aggregator | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cachedapi.inference.net · read 2026-09-16 |
| Vercel AI Gateway aggregator | standard | $5per Mtok in$25per Mtok out$0.5per Mtok cached$6.25per Mtok cache writeai-gateway.vercel.sh · read 2026-08-25 |
| Venice AI aggregator | standard | $6per Mtok in$30per Mtok out$0.6per Mtok cached$7.5per Mtok cache writemodels.dev · read 2026-09-16 |
Measured 67 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 1stof 22 | Cybench — Epoch AI | Unguided % Solved | 0.93 |
| 1stof 41 | Deepresearchbench — Epoch AI | Average score (high) | 0.553 |
| 2ndof 34 | LMArena · Search | Score (Elo) | 1253 |
| 2ndof 402 | LMArena · Text | Score (Elo) (high) | 1505 |
| 2ndof 50 | Metr time horizons — Epoch AI | average_score | 0.789 |
| 3rdof 25 | SWE-Bench Pro (Public Dataset) | Resolve Rate (thinking) | 51.9 |
| 4thof 38 | Gso — Epoch AI | Score OPT@1 (high) | 0.412 |
| 4thof 262 | MMLU-Pro | Overall (accuracy) (thinking) | 89.1 |
| 4thof 47 | SWE-bench Verified (default "Bash Only" view, agent = mini-SWE-agent) | % Resolved | 75.6 |
| 4thof 35 | Swe bench verified — Epoch AI | mean_score | 78.719 |
| 5thof 17 | Cl bench life — Epoch AI | Overall (high) | 0.17 |
| 5thof 152 | LMArena · Vision | Score (Elo) (high) | 1299 |
| 6thof 32 | AIME 2026 | Accuracy (± 95% CI) (high) | 96.67 |
| 6thof 123 | Lmca — Epoch AI | Score (max) | 55.76 |
| 6thof 13 | Rli — Epoch AI | Score | 4.17 |
| 7thof 23 | Cl bench — Epoch AI | Overall | 0.207 |
| 7thof 60 | Vending bench 2 — Epoch AI | Score | 8017.594 |
| 7thof 60 | Vending-Bench 2 — Current leaderboard | Money Balance (average across 5 runs) | 8017.59 |
| 8thof 101 | FrontierMath (Tiers 1-3) | mean_score (max) | 40.7 |
| 8thof 51 | Humanity's Last Exam (Epoch AI replication) | Accuracy (max) | 34.44 |
| 8thof 202 | Terminalbench — Epoch AI | Accuracy mean | 0.798 |
| 10thof 72 | Frontiermath tier 4 — Epoch AI | mean_score (max) | 22.9 |
| 11thof 23 | Gbaeval — Epoch AI | Overall score | 0.441 |
| 11thof 142 | terminal-bench@2.0 | Accuracy | 76.4 |
| 12thof 18 | Algotune — Epoch AI | Score (high) | 1.47 |
| 12thof 180 | SWE-bench Verified (Bash Only toggle off — all agents) | % Resolved | 75.6 |
| 14thof 40 | Design Arena · agenticgamedev (agents) | Elo | 1215 |
| 14thof 21 | Ebr bench — Epoch AI | mean_score (max) | 12.698 |
| 16thof 102 | Simplebench — Epoch AI | Score (AVG@5) | 0.676 |
| 16thof 28 | τ³-Banking Leaderboard | Pass^1 | 27.3 |
| 17thof 55 | Design Arena · androidnative (agents) | Elo | 1218 |
| 18thof 46 | Enigma eval — Epoch AI | Accuracy (max) | 7.6 |
| 18thof 64 | Proofbench — Epoch AI | Accuracy (max) | 50 |
| 18thof 170 | Weirdml — Epoch AI | Accuracy (high) | 77.95 |
| 19thof 63 | Design Arena · mobileapps (agents) | Elo | 1237 |
| 20thof 152 | Design Arena · dataviz (models) | Elo | 1296 |
| 20thof 30 | Frontiercode — Epoch AI | Main score | 0.266 |
| 21stof 64 | Design Arena · fullstack (agents) | Elo | 1228 |
| 22ndof 92 | Design Arena · asciiart (models) | Elo | 1272 |
| 22ndof 109 | Design Arena · svg (models) | Elo | 1253 |
| 22ndof 106 | Frontiermath tiers 1 3 v2 — Epoch AI | mean_score (max) | 65.965 |
| 23rdof 156 | Design Arena · codecategories (models) | Elo | 1304 |
| 23rdof 162 | Design Arena · website (models) | Elo | 1304 |
| 23rdof 161 | Dtbench — Epoch AI | Accuracy (max) | 91.2 |
| 23rdof 121 | Webdev arena — Epoch AI | Arena Score | 1556.29 |
| 24thof 65 | Apex agents — Epoch AI | Pass@1 score | 0.324 |
| 24thof 147 | Design Arena · 3d (models) | Elo | 1302 |
| 24thof 155 | Design Arena · gamedev (models) | Elo | 1301 |
| 26thof 150 | Design Arena · uicomponent (models) | Elo | 1300 |
| 27thof 205 | ARC-AGI-1 (Public Eval) | Score (%) (max) | 96.75 |
| 28thof 62 | Design Arena · webapps (agents) | Elo | 1219 |
| 28thof 128 | LMArena · WebDev | Score (Elo) (high) | 1547 |
| 29thof 80 | Simpleqa verified — Epoch AI | mean_score (max) | 47 |
| 31stof 230 | ARC-AGI-1 (Semi-Private) | Score (%) (high) | 94 |
| 31stof 81 | Forecastbench — Epoch AI | Overall score | 60 |
| 34thof 244 | Arc agi — Epoch AI | Score | 94 |
| 34thof 62 | Frontiermath tier 4 v2 — Epoch AI | mean_score (max) | 26.829 |
| 35thof 216 | ARC-AGI-2 (Public Eval) | Score (%) (high) | 79.03 |
| 37thof 313 | GPQA Diamond | mean_score | 90.53 |
| 39thof 232 | ARC-AGI-2 (Semi-Private) | Score (%) (high) | 69.17 |
| 39thof 127 | Mystery game puzzles — Epoch AI | mean_score (max) | 25 |
| 42ndof 290 | OTIS Mock AIME 2024-2025 | mean_score | 94.444 |
| 44thof 112 | Ale bench — Epoch AI | Performance | 996.5 |
| 46thof 227 | Arc agi 2 — Epoch AI | Score | 69.17 |
| 81stof 105 | Vectara Hallucination Leaderboard | Hallucination Rate | 12.2 |
| 102ndof 222 | Chess puzzles — Epoch AI | mean_score | 17 |
| 129thof 656 | Epoch capabilities index — Epoch AI | ECI Score (max) | 155.25 |
About
Claude Opus 4.6 — a text model from Anthropic, sold by 13 companies from $5 in and $25 out per million tokens, placed 1st of 41 on Deepresearchbench — Epoch AI.
It takes text, images and files and returns text, with a context window of 1,000,000 tokens. It was published in August 2025, trained on material up to May 2025. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning, code and agents. Thirteen companies sell it. The cheapest is $5 in and $25 out per million tokens at AWS Bedrock. Beside the standard rate there are batch, regional and separately routed lanes. It has been measured on 67 boards, and stands best at 1st of 41 on Deepresearchbench — Epoch AI.
Every current figure
- Maker
- Anthropic
- Register
- model
- Takes
- text + image + file
- Returns
- text
- Context
- 1,000,000 tokens
- Published
- August 2025
- Knowledge to
- May 2025
- Licence
- not read
- Sellers
- 13
- Price
- $5 in and $25 out per million tokens — AWS Bedrock
- Boards
- 67
- Best place
- 1st of 41 — Deepresearchbench — Epoch AI
Known as 22 names
Claude-Opus-4.6 (High)Claude Opus 4.6 (120K, High)Anthropic Opus 4.6 (Max)Claude Opus 4.6 (max)Claude Opus 4.6 (no thinking)claude-opus-4-6_maxclaude-opus-4-6-searchclaude-opus-4-6claude-opus-4-6-highClaude-4.6-Opus(Thinking)claude-opus-4-6 (thinking)claude-opus-4-6 (thinking)*Claude 4.6 Opusanthropic.claude-opus-4-6-v1anthropic/claude-opus-4.6anthropic/claude-opus-4.6:batchanthropic/claude-opus-4-6bedrock/claude-opus-4-6vertex/claude-opus-4-6vertex/claude-opus-4-6@europe-west1vertex/claude-opus-4-6@us-east5Anthropic: Claude Opus 4.6 (batch)