Claude Fable 5
anthropic/claude-fable-5claude-fable-5fable-5- π First publicly available model from Anthropic's Mythos class
- π 1M-token context with up to 128K output tokens
- π§ Always-on adaptive thinking for long-horizon agentic work
- ποΈ Multimodal vision capabilities for reasoning over images
- π§ Coding, function calling, and web search built in
- π’ Available via the Claude API and cloud platforms
- π Anthropic reports gains over Opus 4.8 on agentic coding
- π Ships with additional Mythos-class safety safeguards
Anthropic PBC is an American artificial intelligence company headquartered in San Francisco. Structured as a public benefit corporation, the lab develops large language models under the Claude name, with a research emphasis on building reliable, steerable, and safety-focused AIβ¦
Explore 18 more models by Anthropic βClaude Fable 5 is Anthropic's most capable widely released model and the first commercially available member of its top-tier Mythos class, released in 2026.[2][3] Positioned above the Opus tier, it pairs a 1M-token context window with up to 128K output tokens, always-on adaptive thinking, and multimodal vision, and is designed for demanding reasoning and long-horizon agentic work.[2][3]
Compared with its same-family predecessor Claude Opus 4.8 β Anthropic's previous flagship, released just weeks earlier β Fable 5 sits a tier higher and is presented as an advance on agentic coding, reasoning, and visual understanding, with always-on adaptive thinking enabling longer, asynchronous task execution.[1][2] Anthropic reports gains over Opus 4.8 on agentic coding in its announcement, though specific figures should be verified against the company's own model card.[2]
As a Mythos-class model, Fable 5 ships with additional safety safeguards, and its unrestricted counterpart, Mythos 5, is offered through more limited access channels rather than general availability.[2][3] Within Anthropic's broader lineup it succeeds the Opus generation, including Claude Opus 4.7 and Claude Opus 4.6, extending the family's emphasis on coding, tool use, and extended autonomous workflows.[3]
This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.
Daily usageΒ Β·Β Claude Fable 5Β Β·Β Organic providers
Top providersΒ Β·Β Claude Fable 5Β Β·Β Organic
- Venice.ai Proxy41.0%
- β² Apex Ant22.0%
- Quartz15.4%
- ZLKPro-Api6.8%
- GesundAI5.6%
- attacompute.ai3.7%
- Others5.5%
| Provider | Health | Repβ | Sales | Tokens | Routing | Input $/M | Cached $/M | Output $/M | Save | 1st token | tok/s |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Super Seeder 0xd19fβ¦41f3 | healthy | 90.57 | $13.12 | 376.30k | #4 | $9.4268 | $0.9427 | $47.1339 | β6% | β | β |
| β² Apex Ant 0x73b4β¦e736 | healthy | 88.20 | $286.80 | 210.85M | #1 | $5.00 | $0.50 | $25.00 | β50% | β | β |
| Venice.ai Proxy 0x1f22β¦18c9 | healthy | 75.07 | $534.80 | 223.55M | #2 | $5.50 | $0.55 | $27.50 | β45% | β | β |
| π¬ Banter 0x4122β¦e194 | healthy | 74.99 | $0.14 | 7.60k | #3 | $5.00 | $0.695 | $30.00 | β50% | β | β |
| Relay Works 0x3040β¦7139 | β | 58.03 | $0.30 | 107.71k | gated | $6.50 | $0.65 | $32.50 | β35% | β | β |
| ZLKPro-Api 0x0b0bβ¦f446 | β | 37.21 | $88.14 | 225.36M | gated | $3.15 | $1.00 | $15.75 | β69% | β | β |
| Vikey Seller 0xe4c4β¦3c7b | healthy | 30.79 | $0.00 | 230.00 | gated | $5.00 | $1.00 | $25.00 | β50% | β | β |
| Prime Seed 0x71e2β¦fc07 | healthy | 27.20 | $0.01 | 186.00 | gated | $11.00 | $1.10 | $55.00 | β | β | β |
| HallPass 0xcac8β¦71b4 | β | 20.00 | β | β | gated | $1.50 | $0.15 | $7.50 | β85% | β | β |
| klinn 0x0ee1β¦6c41 | β | 12.34 | $0.07 | 31.88k | gated | $5.4031 | $0.5403 | $27.0156 | β46% | β | β |
| Meridian Inference 0x1429β¦fcde | β | 5.10 | β | β | gated | $36.4583 | $36.4583 | $145.8333 | β | β | β |
| dealer 0xc46eβ¦8ce6 | β | 2.11 | β | β | gated | $9.50 | $0.95 | $47.50 | β5% | β | β |
| Hana Gateway β 0x4ae1β¦117b | healthy | 0.51 | $11.10 | 10.54M | gated | $0.45 | $0.05 | $2.00 | β96% | β | β |
| surplus-hive 0x6145β¦5c58 | β | 0.00 | β | β | gated | $14.5833 | $14.5833 | $72.9167 | β | β | β |
| Open Ant 0xe4f6β¦5bc4 | β | 0.00 | $0.03 | 723.00 | gated | $12.00 | $1.20 | $60.00 | β | β | β |
| Fire Ant π₯π 0xbe05β¦bc5d | β | 0.00 | $0.04 | 8.93k | gated | $9.00 | $0.90 | $45.00 | β10% | β | β |
| LLMart 0x948aβ¦a000 | β | 0.00 | β | β | gated | $5.00 | $0.5037 | $25.00 | β50% | β | β |
| Xcat 0xbc33β¦b75f | β | 0.00 | $0.72 | 89.90k | gated | $5.00 | $5.00 | $30.00 | β50% | β | β |
| Joi Gateway 0x8a90β¦7007 | β | 0.00 | β | β | gated | $4.88 | $4.88 | $24.40 | β51% | β | β |
| antseed-neon-puma-944e 0x6650β¦944e | β | 0.00 | $0.71 | 564.73k | gated | $4.092 | $1.20 | $20.46 | β59% | β | β |
| Xpay 0x86b3β¦9438 | β | 0.00 | β | β | gated | $3.40 | $3.40 | $16.00 | β66% | β | β |
| Apex TEE Test 0xe672β¦7955 | β | 0.00 | β | β | gated | $3.24 | $3.24 | $16.20 | β68% | β | β |
| Best 0x472fβ¦69fd | healthy | 0.00 | $17.05 | 60.99M | gated | $3.15 | $2.6775 | $15.75 | β69% | 0.3 s | 145 |
| Open Bird 0xc0f1β¦8183 | β | 0.00 | $4.53 | 1.68M | gated | $1.8998 | $0.19 | $9.499 | β81% | β | β |
| Catastrophic 0xefbfβ¦8908 | β | 0.00 | $4.66 | 12.46M | gated | $0.45 | $0.35 | $1.15 | β96% | β | β |
| Flash 0x0329β¦9b2c | β | 0.00 | βexcluded | β | gated | $0.44 | $0.44not advertised | $1.14 | β96% | β | β |
| antseed-cosmic-gecko-c46d 0xd9c0β¦c46d | healthy | 0.00 | $5.53 | 17.39M | gated | $0.29 | $0.203 | $0.87 | β97% | β | β |
| NovaRoute AI 0xc50dβ¦ed7b | β | 0.00 | $2.40 | 670.93k | gated | $0.2584 | $0.3277 | $0.7752 | β97% | β | β |
"Best price" and the provider table are live Antseed catalog data (advertised $/1M tokens β or $ per generated image for unit-billed image models β not settled amounts). Reputation = buyer trust score (0-100, the Antseed SDK's own formula). "Routing" = the SDK's default buyer routing (what Antseed Desktop ships with): a trust β₯ 60 gate on the effective reputation, then cheapest-first among routable providers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on Antseed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.