Tech Meridian← ALL MODELS
PROMY MERIDIAN

ANTHROPIC · MODEL RELEASE TRACKER

Claude Opus 5

Claude Opus 5 is an Anthropic model announced as available on 2026-09-16. Anthropic describes it as a thoughtful, proactive model that approaches the frontier intelligence of Claude Fable 5 at roughly half the price, with especially strong results on coding and knowledge-work benchmarks and meaningful gains over Opus 4.8. It is the new default on Claude Max and the strongest model on Claude Pro, and Anthropic notes it remains behind Mythos 5 on cybersecurity tasks.

CURRENT SNAPSHOT5/5 DIMENSIONS WITH DATA

The dimensions that change the decision.

PRICING
Input price$5 USD / 1M input tokens; Claude API standard, global routing
Output price$25 USD / 1M output tokens; Claude API standard, global routing
Cached input price$0.5 USD / 1M cached input tokens; Claude API standard, global routing
Estimated cost to label one hour of videoReported estimate of about USD 42 per hour for labeling in the same comparison where Flora was estimated at about USD 0.72/hour and GPT-6 Astra at about USD 88.50/hour.
CONTEXT WINDOW
Context window1,000,000 tokens
MODALITIES
Input modalitiesText, images
Output modalitiesText
BENCHMARKS
Benchmarks (coding & knowledge work)Described as 'the new state-of-the-art' on coding and knowledge-work evaluations such as Frontier-Bench and GDPval-AA.
SWE-Serve inference-engineering evaluationIn NVIDIA’s SWE-Serve evaluation, Claude Opus 5 is listed among top performers with a reported pass@1 of 75% on the evaluated inference-engineering tasks.
Benchmark performance on coding and knowledge-work evaluationsAnthropic reports Opus 5 as the new state-of-the-art on coding and knowledge-work evaluations (e.g., surpasses all models on Frontier-Bench v0.1, more than doubles Opus 4.8’s performance at lower cost per task; on CursorBench 3.2 it reaches within 0.5% of Fable 5’s peak at about half the cost per task; on ARC-AGI 3 its score is ~3× the next-best model; on Zapier AutomationBench its pass rate is ~1.5× the next-best for the same cost; on OSWorld 2.0 it outperforms every other model at a given cost).
Pass@3 (UI-TAU benchmark)Reported as 46.67% Pass@3 on the UI-TAU benchmark (as stated in the cited paper).
Performance on Nvidia's 50-video labeling testReported to have "scored much worse" than Nvidia's Flora (and worse than GPT-6 Astra) in Nvidia's 50-video labeling test.
MARQ median F1 (vs off-the-shelf baseline)MARQ achieved a median F1 of 0.80 for Claude Opus 5 on an expert-labeled NASA requirement set (100 executions), compared with an off-the-shelf baseline F1 of 0.55.
BABA-Bench: rate of reaching required outcomesReached all required outcomes in 81.6% of its runs on the BABA-Bench evaluation.
Participation in 'Into the Unknown' benchmarkClaude Opus 5 was one of nine AI agents evaluated in the Carnegie Mellon–led 'Into the Unknown' benchmark, which asked agents to carry out follow-up research studies on open questions from US national labs.
Co-Design Bench rankingReported to take the next spots after GPT-6 Astra, alongside Qwen3.8-Max, on the Co-Design Bench coding-agent benchmark.
Coding benchmark performanceMentioned as a US model whose coding-benchmark performance level was matched by LongCat-2.5 a few months later.
AVAILABILITY
AvailabilityActive on Claude API, Amazon Bedrock, Google Cloud, Microsoft Foundry and Claude Platform on AWS. Active (legacy)

RELEASE TIMELINE

Published, source-backed release events only.

1 SOURCES · IMPORTANCE 7.0

Cloudflare launches Auto Router public beta in AI Gateway to rate requests by stakes and other metrics

Cloudflare published Auto Router in public beta via AI Gateway: a two-stage router that builds a candidate model pool then runs a multi-head classifier on Workers AI to assign probabilities across 14 task categories and score requests 1–5 on complexity, ambiguity, stakes and context dependence, using only the newest turns (up to 8 KiB). The router combines those scores with benchmark results (quoted formula: "utility = expected quality - adaptive cost penalty"), exposes routing headers for auditability, supports per-request allowed-models headers, is free in beta, and currently lacks WebSocket support.

→
1 SOURCES · IMPORTANCE 7.0

Nvidia open-sources Flora, a Qwen3.5-based robot video labeling engine

Nvidia's GEAR lab open-sourced Flora, a data-engine composed of three fine-tuned Qwen3.5 models (9B for segmentation, 27B for descriptions, and a 27B 'Flora-Rewarder' for scoring) that labels robot videos for action boundaries, step-level rewards, and task success. Nvidia says Flora is already used in training its GR00T models at scale on GB300 GPUs and estimates labeling costs of about $0.72/hour versus ~$88/hour with GPT-6 Astra, though the cost figures are projections and benchmark comparisons include methodological caveats.

→
1 SOURCES · IMPORTANCE 6.0

Deepseek releases Deepseek‑V4.1‑Flash‑1017, open-sources weights and opens search APIs

Chinese company Deepseek released Deepseek‑V4.1‑Flash‑1017 and published the model weights on Hugging Face and ModelScope. The company says the model—trained with reinforcement learning using self-feedback—runs faster, produces shorter outputs, beats larger rivals in its internal coding and agent benchmarks, and is paired with new Search & Fetch APIs for developers; independent verification of the benchmark claims is not yet available.

→
6 SOURCES · IMPORTANCE 8.0

Anthropic releases Claude Opus 5 — lower-cost model claiming near‑frontier performance

Anthropic announced Claude Opus 5, a new model positioned as a cost‑efficient successor to Opus 4.8 and the new default on Claude Max (and the strongest on Claude Pro). Anthropic says Opus 5 matches or exceeds prior models on many coding, knowledge‑work and scientific benchmarks (Frontier‑Bench, GDPval‑AA, CursorBench, ARC‑AGI, Zapier AutomationBench, OSWorld) at lower cost per task while remaining behind Mythos 5 on security and biology frontier tasks; the company also reports improved alignment and safety in pre‑deployment audits and links a System Card for more details.

→

VERIFIABLE FACTS

Every value stays attached to a source and date.

OTHER · Performance vs Claude Fable 5DEVELOPER CLAIM

Described as coming close to the frontier intelligence of Claude Fable 5 while costing about half as much.

BENCHMARKS · Benchmarks (coding & knowledge work)DEVELOPER CLAIM

Described as 'the new state-of-the-art' on coding and knowledge-work evaluations such as Frontier-Bench and GDPval-AA.

BENCHMARKS · Benchmark performance on coding and knowledge-work evaluationsDEVELOPER CLAIM

Anthropic reports Opus 5 as the new state-of-the-art on coding and knowledge-work evaluations (e.g., surpasses all models on Frontier-Bench v0.1, more than doubles Opus 4.8’s performance at lower cost per task; on CursorBench 3.2 it reaches within 0.5% of Fable 5’s peak at about half the cost per task; on ARC-AGI 3 its score is ~3× the next-best model; on Zapier AutomationBench its pass rate is ~1.5× the next-best for the same cost; on OSWorld 2.0 it outperforms every other model at a given cost).

LIMITATIONS · Sensitivity to scientific prose and biological priorsINDEPENDENTLY SUPPORTED

The arXiv study reports that in scientific prose Opus 5 often prefers the biologically expected assignment; when that biological preference is removed, recovery of the better-supported assignment increases from 27% to 79%, and reaches 92% when the Biology-favored record is accompanied by a formalization request and explicit paired-design cue.

WHAT CHANGED

Stored passport versions, without reconstructed history.

Passport updated31 facts
BENCHMARKS · Coding benchmark performance+ Mentioned as a US model whose coding-benchmark performance level was matched by LongCat-2.5 a few months later.
Passport updated30 facts
BENCHMARKS · Co-Design Bench ranking+ Reported to take the next spots after GPT-6 Astra, alongside Qwen3.8-Max, on the Co-Design Bench coding-agent benchmark.
Passport updated29 facts
BENCHMARKS · Participation in 'Into the Unknown' benchmark+ Claude Opus 5 was one of nine AI agents evaluated in the Carnegie Mellon–led 'Into the Unknown' benchmark, which asked agents to carry out follow-up research studies on open questions from US national labs.
Passport updated28 facts
BENCHMARKS · BABA-Bench: rate of reaching required outcomes+ Reached all required outcomes in 81.6% of its runs on the BABA-Bench evaluation.
SAFETY · BABA-Bench: procedural compliance / pass rate+ Only 52.5% of its runs passed all checks that verify required procedures (i.e., procedural compliance) in the BABA-Bench evaluation.
Passport updated26 facts
CAPABILITIES · Defect detection and false-alarm control+ In the MARQ evaluation, Claude Opus 5 improved defect detection and false-alarm control compared with other evaluated models.
LIMITATIONS · Gains concentrated in Correctness and Conformance criteria+ Criterion-level analysis reported in MARQ showed that gains were concentrated in Correctness and Conformance, indicating model-dependent responses across several criteria.
BENCHMARKS · MARQ median F1 (vs off-the-shelf baseline)+ MARQ achieved a median F1 of 0.80 for Claude Opus 5 on an expert-labeled NASA requirement set (100 executions), compared with an off-the-shelf baseline F1 of 0.55.
Passport updated23 facts
BENCHMARKS · Performance on Nvidia's 50-video labeling test+ Reported to have "scored much worse" than Nvidia's Flora (and worse than GPT-6 Astra) in Nvidia's 50-video labeling test.
PRICING · Estimated cost to label one hour of video+ Reported estimate of about USD 42 per hour for labeling in the same comparison where Flora was estimated at about USD 0.72/hour and GPT-6 Astra at about USD 88.50/hour.
PRICING · Pricing and cost-efficiency− Anthropic positions Opus 5 as delivering greatly improved performance for the same cost as its predecessor Opus 4.8 and says it comes close to the intelligence of Claude Fable 5 at about half the price.
Passport updated22 facts
BENCHMARKS · Pass@3 (UI-TAU benchmark)+ Reported as 46.67% Pass@3 on the UI-TAU benchmark (as stated in the cited paper).
Passport updated21 facts
BENCHMARKS · Benchmark performance on coding and knowledge-work evaluations+ Anthropic reports Opus 5 as the new state-of-the-art on coding and knowledge-work evaluations (e.g., surpasses all models on Frontier-Bench v0.1, more than doubles Opus 4.8’s performance at lower cost per task; on CursorBench 3.2 it reaches within 0.5% of Fable 5’s peak at about half the cost per task; on ARC-AGI 3 its score is ~3× the next-best model; on Zapier AutomationBench its pass rate is ~1.5× the next-best for the same cost; on OSWorld 2.0 it outperforms every other model at a given cost).
LIMITATIONS · Relative performance on cybersecurity tasks+ Anthropic states Opus 5 remains behind Mythos 5 on cybersecurity tasks.
PRICING · Pricing and cost-efficiency+ Anthropic positions Opus 5 as delivering greatly improved performance for the same cost as its predecessor Opus 4.8 and says it comes close to the intelligence of Claude Fable 5 at about half the price.
PRICING · Pricing (relative)− Reported to cost about half the price of Claude Fable 5.
CAPABILITIES · Recovery of best-supported assignment when constraints are formalized+ In a controlled research task, when constraints were stated directly, Claude Opus 5 recovered the best-supported assignment in 96% of cases (reported by the arXiv study).
LIMITATIONS · Sensitivity to scientific prose and biological priors+ The arXiv study reports that in scientific prose Opus 5 often prefers the biologically expected assignment; when that biological preference is removed, recovery of the better-supported assignment increases from 27% to 79%, and reaches 92% when the Biology-favored record is accompanied by a formalization request and explicit paired-design cue.
BENCHMARKS · SWE-Serve inference-engineering evaluationListed among top performers on SWE-Serve; a mini-swe-agent evaluation reported Claude Opus 5 at 75% mean pass@1 in the referenced tasks.→In NVIDIA’s SWE-Serve evaluation, Claude Opus 5 is listed among top performers with a reported pass@1 of 75% on the evaluated inference-engineering tasks.