Anthropic releases Claude Sonnet 5.5 with faster speeds and lower task costs
Anthropic launched Claude Sonnet 5.5, advertising over 30% faster generation speeds and up to 30% lower per-task costs than its predecessor.
Anthropic has launched Claude Sonnet 5.5, positioning the mid-tier model as a faster and more cost-effective option that outputs answers over 30% faster than Sonnet 5 while reducing per-task costs by up to 30%. CNBC reported the launch is Anthropic's second release since its chief executive called for a slowdown in AI development. The new model is the second entry in the Claude 5.5 family following the rollout of Opus 5.5, with Haiku 5.5 planned for release in the coming weeks.[6][13][12][9][10][2]
Sonnet 5.5 retains Sonnet 5's token pricing at $2 per million input tokens and $10 per million output tokens, which Decrypt noted is half the token cost of Opus 5.5. On benchmarks, Sonnet 5.5 scored 70.6% on Terminal-Bench 4.0 compared to 10.3% for Sonnet 5, outscoring Opus 5.5 on agentic coding. Although Anthropic attributes per-task cost savings to using fewer tokens for identical tasks, an independent tester cited by Decrypt found the model burned more tokens than any model it had previously measured.[12][8][5][4][7][10]
Available across Claude, AWS, Google Cloud, and Microsoft Azure, the model is targeted at everyday coding, bug fixing, and document creation. The New Stack reported that Sonnet 5.5 is the first Sonnet release to include Anthropic's three-stage cyber classifiers and model fallbacks. Additionally, Digit reported that Anthropic claims Claude Sonnet 5.5 beats GPT-6 Sol at long-horizon knowledge work.[12][11][1]
Key facts
- Anthropic launched Claude Sonnet 5.5 as the second model in the Claude 5.5 model family.
- The model runs over 30% faster than Sonnet 5 and cuts per-task costs by up to 30%.
- Per CNBC, the launch represents Anthropic's second release since its CEO called for a slowdown in AI development.
- Sonnet 5.5 pricing is $2 per million input and $10 per million output tokens, half the price per token of Opus 5.5.
- Sonnet 5.5 scored 70.6% on Terminal-Bench 4.0 compared to 10.3% for Sonnet 5, surpassing Opus 5.5 in agentic coding.
- An independent tester cited by Decrypt reported that Sonnet 5.5 burned more tokens than any model it had measured.
- Sonnet 5.5 is the first Sonnet model to feature three-stage cyber classifiers and model fallbacks.
- Anthropic plans to release Haiku 5.5 in the coming weeks.
Sources · 12 sources
- DI
Digit@digitindiaPost on X ·
Claude Sonnet 5.5 is here, and Anthropic says it beats GPT-6 Sol at long-horizon knowledge work. Here’s everything you need to know 👇 https://t.co/unxGpjjwKj
Open source - NE
NeowinFeed@NeowinFeedPost on X ·
Positioned directly below the recently announced Claude Opus 5.5, the mid-tier Sonnet 5.5 model is over 30% faster than Sonnet 5 while nearly matching flagship-grade benchmarks. #Anthropic #Sonnet5.5 https://t.co/RUdhE8yYgT
Open source - TR
TrendSpider@TrendSpiderPost on X ·
BREAKING: Anthropic has released Sonnet 5.5 It runs 30%+ faster than Sonnet 5 and costs up to 30% less for “most work”. https://t.co/hLWKeopshh
Open source - PM
Polymarket Money@PolymarketMoneyPost on X ·
JUST IN: Sonnet 5.5 just beat Opus 5.5 on Terminal Bench 4.0, taking the lead in “agentic coding.”
Open source - DE
DecryptArticle ·
Anthropic's Claude Sonnet 5.5 Is Out, Beats Opus 5.5 at Coding for Half the Price Anthropic's mid-tier model tops its own flagship on Terminal-Bench 4.0 and costs half as much per token, but an independent tester found it burns more tokens than any model it has measured.
Open source - EV
Evan@StockMKTNewzPost on X ·
ANTHROPIC JUST SAID THAT IT IS - RELEASING SONNET 5.5, SECOND MODEL IN CLAUDE 5.5 MODEL FAMILY
Open source - WS
Wall St Engine@wallstenginePost on X ·
Sonnet 5.5 beats Opus 5.5 in Agentic coding (Terminal Bench 4.0) 😮 https://t.co/DBHYkExG8c
Open source - CB
Crypto Briefing@Crypto_BriefingPost on X ·
🤖NEW: Anthropic launches Claude Sonnet 5.5, a faster and cheaper model for coding, everyday work and document creation. Anthropic says it generates output more than 30% faster than Sonnet 5 and costs up to 30% less per task, while pricing remains $2/M input and $10/M output tokens.
Open source - TE
Techmeme@TechmemePost on X ·
Anthropic releases Sonnet 5.5, saying it generates outputs 30%+ faster than Sonnet 5 and costs up to 30% less per task, and plans to release Haiku 5.5 soon (Anthropic) (Visit Techmeme dot com for the link and full context!)
Open source - AG
AGTP@AGTPinsightsPost on X ·
Claude Sonnet 5.5 just launched, running over 30% faster than Sonnet 5 while needing far fewer tokens to complete the same tasks. It's the second model in the Claude 5.5 family, following Opus 5.5 which launched on September 22nd. Claude Haiku 5.5 is expected to join the lineup in the coming weeks. Sonnet 5.5 is positioned as a faster, cheaper complement to Opus 5.5, built for well scoped everyday tasks like fixing bugs, writing documents, and polishing slides and spreadsheets. On several benchmarks, Sonnet 5.5 at low or medium effort beats Sonnet 5's best score for roughly a tenth of the cost. Some users have also reported it outperforming Opus 5.5 on certain agentic coding and terminal benchmarks. Pricing remains unchanged from Sonnet 5, and the model is available everywhere starting today.
Open source - TN
The New Stack@thenewstackPost on X ·
Claude Sonnet 5.5 is the first Sonnet with Anthropic's three-stage cyber classifiers and model fallbacks, and that changes how developers build on it. https://t.co/oodzgKiBIC
Open source - WS
Wall St Engine@wallstenginePost on X ·
ANTHROPIC RELEASES SONNET 5.5 Sonnet 5.5 is 30%+ faster than Sonnet 5 and costs up to 30% less per task, while keeping the same token pricing: Sonnet 5.5 vs Opus 5.5 pricing: Input: $2/M vs $4/M Output: $10/M vs $20/M Cache writes: $2.50/M vs $5/M Cache reads: both $0.20/M So Sonnet 5.5 is basically 50% cheaper than Opus 5.5 on input, output and cache writes. The efficiency gain comes from using fewer tokens to complete the same work. Anthropic positions Sonnet 5.5 as the faster, lower-cost complement to Opus 5.5, focused on everyday work, coding, bug fixing and creating documents, slides and spreadsheets. On benchmarks: Terminal-Bench 4.0: 70.6% vs 10.3% for Sonnet 5 CursorBench: 55.5% vs 34.1% GDPval-AA: 1844 vs 1846 for Opus 5.5 Anthropic says Opus 5.5 still remains stronger for complex, open-ended work requiring sustained judgment. Sonnet 5.5 is available across Claude, AWS, Google Cloud and Microsoft Azure. Anthropic says Haiku 5.5 is coming in the next few weeks for faster, high-volume and cost-sensitive workloads.
Open source - CN
CNBC@CNBCPost on X ·
Anthropic launches cheaper AI model, its second release since CEO's call for a slowdown https://t.co/JZCUWBUxwP
Open source