Anthropic has released Claude Fable 5.2, taking the #1# spot on the Artificial Analysis Intelligence Index v5 at 65, led by gains in science, knowledge, and coding At max effort it scores 65, 5 points ahead of Gemini 4 Carbon (high, 60) and 6 ahead of Claude Opus 5.5 (max, 59). Gains are driven by scientific reasoning, knowledge and FrontierCode 2.0, and roughly matching Opus 5.5 on AA-Briefcase and Terminal-Bench 4.0. Priced at $8/$40 per 1M input/output tokens, a 20% reduction from Fable 5.1, Claude Fable 5.2 averages a Cost per Task of $26.5 with max effort, a ~2.6x increase compared to Opus 5.5 (max, $10.0) and ~2x compared to Fable 5.1 (max, $12.99) Key takeaways: ➤ Claude Fable 5.2 leads 5 of the 9 evaluations in the Intelligence Index: At max effort, Fable 5.2 scores 65, ahead of Gemini 4 Carbon (high, 60), Claude Opus 5.5 (max, 59) and GPT-6 Astra (max, 57), and ~15 points above Claude Fable 5.1 (max, 49). It sets the highest scores on HLE-Diamond (80%, previous best 69% from Gemini 4 Carbon), AA-Omniscience v1.1 (67, previous best 54 from Gemini 4 Carbon), Terminal-Bench-Science (73%, previous best 65% from Gemini 4 Carbon), CritPt v2 hard subset (77%, previous best 72% from GPT-6 Astra) and FrontierCode 2.0 (49%, previous best 46% from Claude Opus 5.5) ➤ Less progress in agentic knowledge work and terminal coding: Claude Fable 5.2 improves on Fable 5.1 across capabilities, but sometimes only to Opus 5.5 levels. It reaches 1785 Elo on AA-Briefcase, within the confidence interval of Opus 5.5 (1800) and just behind Claude Sonnet 5.5 (1813). On GDP.pdf it reaches 24% all-pass against 26% for Opus 5.5. On Terminal-Bench 4.0 it scores 60%, level with Opus 5.5 and behind Kimi K3.1 Pro (67%), Gemini 4 Carbon (65%) and Sonnet 5.5 (64%). AutomationBench-AA is an exception, at 76% against 70% for Opus 5.5, narrowly behind Gemini 4 Carbon, 78% ➤ Lower efforts still take leading positions: With high and xhigh effort, Claude Fable 5.2 scores 62, ahead of every other model at any effort, using 98k output tokens per task against 173k for Opus 5.5 (max). At max effort Fable 5.2 costs ~2.6x more than Opus 5.5, but at lower effort settings it can outperform Opus 5.5 at lower cost per task Other model details: ➤ Context window: 1 million tokens with image and text input ➤ Pricing: $8/$40 per 1M input/output tokens; cache writes $10 with 5-minute TTL, $16 with 1-hour TTL, and cache reads $0.20 ➤ Effort settings: five (low, medium, high, xhigh, max). Intelligence Index evaluations were run at all five with Anthropic's default fallback enabled Congratulations to @AnthropicAI on this impressive release!

We've launched our HLE-Diamond leaderboard, our independent evaluation of the refined 1,000-question version of Humanity's Last Exam, as part of v5 of the Artificial Analysis Intelligence Index HLE-Diamond was released by @CAIS and @scale_AI after a year of cleaning and refining the original Humanity's Last Exam dataset with input from the research community. It has 1,000 closed-ended questions, split evenly between reasoning and knowledge, and 272 of them include an image. We run every question once without tools, using the official HLE system prompt. GPT-5.6 Luna (medium) grades each answer against the reference answer with the official HLE judge prompt, and we report the share of questions judged correct (pass@1). Key takeaways: ➤ Only 17 model variants (~13%) score above 50% today: Gemini 4 Carbon (high) leads at 69%, followed by Claude Opus 5.5 (max) at 68% ➤ Models score higher on reasoning than on knowledge: Over 90% of the models we measured do better on the reasoning portion of the benchmark. Gemini 4 Carbon (high) leads both, at 80% on reasoning and 59% on knowledge. Claude Opus 5.5 (max) and GPT-6 Astra (max) both score 78% on reasoning, and Opus 5.5 is ahead on knowledge, 58% vs 50% ➤ Open weights models trail by more than 30 points: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 37%, followed by MiniMax-M3.1-Flash at 28% and Kimi K3 (max) at 27%

We’ve announced the Artificial Analysis Progress Index: a single capability scale for every language model since 2022, launched alongside Intelligence Index v5 The Intelligence Index is still our primary measure of current model intelligence. It uses the evaluations most relevant to developers today, so its mix changes as evaluations are added and retired, and scores from different versions can’t be compared. The Progress Index is built for comparisons across time. It uses item response theory, a statistical method used to score standardized tests, to estimate one general capability score for each model from the underlying tasks in all our evaluations, including those that are retired or excluded from the Intelligence Index. Use the Intelligence Index to choose a model today. Use the Progress Index to see how capability has changed over time. The Progress Index builds on prior work placing models on a single capability scale, in particular @EpochAIResearch's Epoch Capabilities Index and the capability scale @AnthropicAI published in its Claude Mythos Preview system card. Key takeaways: ➤ Since reasoning models arrived, the frontier has risen ~65 to 70 points a year, about double the 2023 pace: the highest score among released models rose 35 points in 2023 (GPT-3.5 Turbo 125 to GPT-4 Turbo 160), 67 in 2024 (to o1 at 228), 68 in 2025 (to GPT-5.2 at 296) and 56 so far in 2026 (to Gemini 4 Carbon at 352). ➤ Open weights models generally trail proprietary models by about 4 months: since January 2025, each new highest open weights score came 2-8 months (median ~4) after a proprietary model first reached it. DeepSeek V4.1 Flash 1017 is the highest-scoring model with released weights at 324, a level Claude Fable 5 reached in June 2026. Kimi K3.1 Pro (weights not yet released) scores 341, about two months after the closed frontier ➤ New frontier scores are being set more often: the highest score among released models was raised on 2 release dates in 2023, 4 in 2024, 9 in 2025 and 11 so far in 2026. That is about once every four weeks this year. ➤ OpenAI has held the top score most of the time since November 2022, but the frontier has become a close contest: OpenAI held the number one position for the first 27 months measured, but in 2026 especially there have been rapid changes in the leader between OpenAI, Anthropic, and now Google

Phonic's Phonic 1 (High) is the new leading Speech to Speech model, debuting at #1# on the Artificial Analysis Speech to Speech Index at 87.1, while also leading on agentic performance and being competitive on latency Phonic 1 is @phonic_co 's Speech to Speech model, available through the Phonic API at $0.07 per minute. We evaluated Phonic 1 at its high intelligence level through the Phonic API. Key takeaways: ➤ Speech to Speech Index: #1# of 22 at 87.1, ahead of Grok Voice 3.0 (86.9) by a very small margin. The Index averages Speech Reasoning (Big Bench Audio), Agentic Performance (Tau Voice), Arena Preference and Task Success Rate ➤ Speech Agent Arena: #8# of 30 on preference (Elo 1012) and #4# on Task Success Rate (92.3%) ➤ Tau Voice: #1# at 79.9%, a clear lead of 5.4 points over the next model ➤ Big Bench Audio: #3# at 98.9%, behind StepAudio 3 Realtime (99.7%) and Qwen Audio 3.0 Realtime Plus (99.2%) ➤ Speed: #4# on Time to First Audio at 0.61s, and the fastest of the top 5 models in the Index ➤ Cost: $4.20 per hour of input audio Congrats to the @phonic_co team 🚀 See below for more detail ⬇️

Mistral Large 4 is now open weights, narrowly making it the #1# open weights model developed outside of China @MistralAI has updated and released the weights of their flagship model, Mistral Large 4, with 1.05T total and 52B active parameters. With high reasoning, Mistral Large 4 scores 26 on the Artificial Analysis Intelligence Index v5, up 1 point from Mistral Large 4 Preview (25), with the major differences being increased token efficiency and reduced hallucinations. It sits 1 point ahead of the leading US-developed open weights model, Muse Spark Open (xhigh, 25). On the Artificial Analysis Cyber Index, Mistral Large 4 (high) scores 51, third among current open weights models behind DeepSeek V4.1 Flash 1017 (max, 57) and MiMo-V2.6-Pro (56). Key takeaways: ➤ Narrowly leads non-Chinese open weights models: Mistral Large 4 (high) leads Muse Spark Open (xhigh), the next-best open weights model developed outside China, with the largest gap on Terminal-Bench 4.0 (25% vs. 8%). It trails leading open weights models DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) by 6 points on the Intelligence Index, largely due to weaker coding and agentic knowledge work performance: 15% on FrontierCode 2.0 (vs. 29% and 22%, respectively) and 1349 Elo on AA-Briefcase (vs. 1570 and 1504) ➤ Increased token efficiency: Mistral Large 4 (high) uses 82K output tokens per Intelligence Index task, 39% fewer than Preview (135K), placing it on the open weights Pareto frontier. The seven open weights models that score higher use between 113K and 225K output tokens per task ➤ #2# open weights model on the Coding Agent Index: On the Artificial Analysis Coding Agent Index, Mistral Large 4 (high) scores 48 in Mistral Vibe while using 8.5M tokens per task, about half as many as the leading open weights model GLM-5.3 (max) in Opencode (49, 15.9M). It also scores above another leading open weights model, DeepSeek V4.1 Flash 1017 (max), in DeepSeek Harness (46, 16.7M) ➤ Reduced hallucinations: Mistral Large 4 (high) has a 28% hallucination rate on the updated AA-Omniscience v1.1, down from 41% for Preview and matching Muse Spark Open (xhigh). However, its accuracy remains relatively low at 26%, similar to Ling 3.1 Flash 1019 (27%), and behind DeepSeek V4.1 Flash 1017 (max, 40%) and Muse Spark Open (xhigh, 42%) ➤ Lower price and cost per task than the Preview: Mistral Large 4 is priced at $0.70/$2.10 per 1M input/output tokens, a ~50% reduction from the Preview's pricing of $1.36/$4.18. At the new pricing, it costs $2.06 per Intelligence Index task, down 63% from $5.54 for the Preview. This is comparable to Muse Spark Open (xhigh, $2.02) but more costly than leading open weights models DeepSeek V4.1 Flash 1017 (max, $0.60) and MiMo-V2.6-Pro ($0.26) Key model details: ➤ Size: 1.05T total parameters / 52B active parameters ➤ Context window: 1M tokens ➤ Multimodality: Text and image input, text output ➤ Pricing: $0.70/$2.10 per 1M input/output tokens ($0.07 per 1M cached input tokens) ➤ License: MEOW License (Mistral Ecosystem Open-Weight License), where operating a Model-as-a-Service business requires a separate commercial license with Mistral Congratulations to the team at @MistralAI on the release!

Same robot, same task, two models with very different writing styles ➤ On the left, Claude Opus 5.5 dabs out "hello world" dot by dot: 3,270 arm moves over 34 minutes ➤ On the right, GPT-6 Astra writes in continuous strokes: 514 moves in 13 minutes From AA-Robotics v0.1, seen from the model’s view See more examples here:

In AA-PersonalityBench, Meta AI’s Muse models are the most Friendly of all models The Friendly trait measures whether a response conveys warmth and approachability without feeling performative. ➤ Most Friendly: Muse Spark 1.3 (100), Muse Spark Open (98), Muse Wonder (95) and Muse Glimmer (93). Congratulations @ alex wang! ➤ Least Friendly: Nemotron 3 Ultra (1), GPT-3.5 Turbo (3), MiniMax-M3.1-Flash (6), Grok 4.6 (8) and DeepSeek V4.1 Flash 1017 (11)

Google’s Gemini 4 Carbon is the new #1# on the Artificial Analysis Intelligence Index v5, and second only to Claude Opus 5.5 on the Artificial Analysis Coding Agent Index Gemini 4 Carbon scores 59 on Intelligence Index v5, driven by leading performance in complex reasoning, knowledge, and scientific tasks. It also has large gains in coding and knowledge work, particularly when compared to other Gemini models, but it is not leading in all these areas. As with launch of Gemini 4 Argon, Gemini 4 Carbon is being rolled out to selected users and is not yet publicly available. Google is offering a 50% launch discount, with no confirmed end date. All our charts and figures use standard pricing for both Gemini 4 Carbon and Argon. Key results for Gemini 4 Carbon with high reasoning: ➤ Gemini 4 Carbon leads the Intelligence Index v5, with the #1# score in 4 of the 9 Index evaluations: HLE-Diamond, Terminal-Bench-Science, AA-Omniscience, AutomationBench-AA. Carbon makes substantial gains elsewhere but is not leading in coding and knowledge work ➤ Behind only Claude Opus 5.5 on coding agents: With Antigravity CLI, Gemini 4 Carbon (high) scores 59 only 3 points behind Claude Code with Claude Opus 5.5 (max, 62) on the Artificial Analysis Coding Agent Index. Carbon ranks second on both Terminal-Bench 4.0 (64%) and FrontierCode v1.1 (51%), behind Claude Opus 5.5 (max) at 56%. Google has historically underperformed in this area and so this improvement stands out ➤ Improved agentic knowledge work, but not yet leading: On AA-Briefcase, Gemini 4 Carbon (high) scores 1609 Elo, a 126-point improvement over Gemini 4 Argon (high, 1483), but still behind models including Grok 4.7 (xhigh, 1638), Muse Wonder (max, 1659) and Kimi K3.1 Pro (1678). The largest gain comes from Presentation, up 293 Elo to 1598, alongside a 120-point improvement in Analytical Quality to 1671 ➤ Lower Cost per Task than Claude Opus 5.5 for comparable overall intelligence: Gemini 4 Carbon costs $7.76 per Intelligence Index task, ~23% less than Claude Opus 5.5 (max, $10.04) but ~1.2x GPT-6 Astra (max, $6.25). It uses 123k output tokens per task, 29% fewer than Claude Opus 5.5 (max, 173k). At the same $4/$20 standard pricing, its Cost per Task is in line with Gemini 4 Argon ($7.68), while scoring 9 points higher on the Intelligence Index ➤ Leading AA-Omniscience v1.1 score, with low hallucination: Gemini 4 Carbon scores 54 on our recently updated AA-Omniscience v1.1, 6 points ahead of Claude Opus 5.5 (max, 48) and 8 points ahead of Gemini 4 Argon (46). Accuracy (% correct) rises 10 points over Argon to 58%, while its hallucination rate rises from 4% to 11%. This is still the second lowest of any model, behind only Argon, compared with 50% for both Claude Opus 5.5 (max) and GPT-6 Astra (max) Key model details: ➤ Context Window: 1M tokens ➤ Multimodality: Text, image, video, and speech input, with text output ➤ Pricing: $4/$20 per 1M input/output tokens, the same as Gemini 4 Argon’s standard pricing, with a 50% discount. Cached input tokens receive a 95% discount ($0.20 per 1M) Congratulations to the @GoogleDeepMind team!

GitHub has released Copilot Fusion in GitHub Copilot, which orchestrates between Anthropic and OpenAI models. It scores 2 points below Opus 5.5 in Claude Code at half the Cost per Task in the Coding Agent Index GitHub describes Copilot Fusion as a runtime orchestration system, which selects and coordinates models to optimize performance, cost, and latency for each task. Copilot Fusion chooses between a number of models from Anthropic and OpenAI, including Claude Opus 5.5, Claude Sonnet 5.5, GPT-6 Astra, and GPT-6 Sol. We benchmarked its max setting in GitHub Copilot CLI in the Artificial Analysis Coding Agent Index v2.0. Token costs are charged based on the underlying models used to complete a request. Key takeaways: ➤ Cost-efficient for near-frontier performance: Copilot Fusion in GitHub Copilot costs $5.59 per task, versus $11.42 for Claude Code with Opus 5.5 (max). Devin Fusion with Opus 5.5 (max) + GPT-6 Luna (high) scores 62 at $8.00 per task. ➤ Slight lead in Terminal-Bench 4.0: The Coding Agent Index v2.0 weights its three constituent benchmarks equally. Copilot Fusion in GitHub Copilot leads Terminal-Bench 4.0 with a score of 71. It ranks third on FrontierCode in the chart at 51, behind Claude Code with Opus 5.5 (max) at 56 and Devin Fusion with Opus 5.5 (max) + GPT-6 Luna (high) at 54. Its SWE-Atlas-QnA-AA score lags at 57, between GPT-6 Astra in Codex and Gemini 4 Argon in Antigravity CLI. ➤ Half the token use of Opus 5.5 in Claude Code: GitHub Copilot Fusion used 7.2M tokens per task in the Coding Agent Index, approximately half of the tokens used by Claude Opus 5.5 in Claude Code, and 30% less than Devin Fusion with Opus 5.5 (max) + GPT-6 Luna (high).

To measure speed and cost efficiency, we benchmarked Nemotron 3.5 Transcribe on DeepInfra's hosted endpoint. On DeepInfra, it reaches a Speed Factor of 75x, transcribing 1 minute of audio in under 1 second, and costs $2.00 per 1,000 minutes of audio.

NVIDIA's Nemotron 3.5 Transcribe ties for #1# among open weights Speech to Text models with 2.8% WER, using just 4B active parameters Nemotron 3.5 Transcribe is NVIDIA's new open weights Speech to Text model for non-streaming transcription, with 30B total and 4B active parameters, and the most accurate NVIDIA model on our non-streaming Speech to Text leaderboard. Key takeaways ➤ Accuracy: Nemotron 3.5 Transcribe scores 2.8% on the AA-WER Index, tied with Mistral's Voxtral Small for #1# among open-weight models. It leads Voxtral Small on AA-AgentTalk, our dataset of voice agent conversations, at 1.8% vs. 2.3%. ➤ Model size: With 4B active parameters (30B total), Nemotron 3.5 Transcribe sits on the open weights frontier of AA-WER vs. active parameters. It matches the accuracy of Mistral's Voxtral Small with 6x fewer active parameters (4B vs. 24B), while other open weights models of a similar size sit at around 4% WER. Congratulations to @NVIDIAAI on the release! See more details below ⬇️

Open weights models fall further behind in v5. The leader, Kimi K3.1 Pro (weights not yet released), scores 46, 12 points behind the leader, compared with a 3-point gap in v4.3. Of models with released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) score highest at 32, and Claude Haiku 5.5 (max) outscores all of them. The gap is widest on HLE-Diamond and Terminal-Bench-Science, where Kimi K3.1 Pro scores 37% and 34%, compared with 68% for Claude Opus 5.5 (max) and 63% for GPT-6 Astra (max).

Announcing Artificial Analysis Intelligence Index v5, featuring 2 new benchmarks and 3 upgrades. For the first time, the majority of the Index is built on private datasets The Artificial Analysis Intelligence Index is our composite measure of model capabilities across tasks reflecting real-world usage today. v5 is our biggest upgrade since v4 launched in January. With v5, 60% of the Index is now based on evaluations with private test sets or held-out answers. Alongside Intelligence Index v5, we’re launching the Artificial Analysis Progress Index v1, which tracks model capability over time on one continuous scale. New to the Intelligence Index: ➤ FrontierCode 2.0: Cognition's newly-launched private benchmark of real-world coding tasks, graded on whether a repository maintainer would merge the agent's change ➤ Terminal-Bench-Science: 70 agentic scientific research tasks, run in a terminal environment Upgraded: ➤ HLE-Diamond (replacing Humanity's Last Exam): a refined 1,000-question version of Humanity's Last Exam, split evenly between reasoning and knowledge questions ➤ AA-Omniscience v1.1: An updated dataset with saturated and public set questions removed, new questions added, and ambiguous or erroneous questions revised ➤ CritPt v2: a private, 40-question hard subset of CritPt v2, revised by the CritPt authors and run with a code-execution tool Staying in the Index: AA-Briefcase, AutomationBench-AA, GDP.pdf and Terminal-Bench 4.0 Removed from the Index: GDPval-AA v2.1, SciCode, and AA-LCR v1.1 Key results: ➤ Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55): Anthropic and OpenAI hold the top five models. Gemini 4 Argon (high) is next at 50 ➤ The open weights gap widens from 3 points to 12: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 46. Among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) lead at 32 - 26 points behind the leader and below Claude Haiku 5.5 (max, 36) ➤ Cost per task varies by more than 13x among the leading models: of the 10 highest-scoring models shown, GPT-6.1 Terra (max) is the cheapest at $0.96 per Intelligence Index task and Claude Fable 5.1 (max) the most expensive at $13.04

Introducing Code2Video-AA, our independent implementation of HeyGen’s Code2Video Bench, which evaluates how well models create product and brand marketing videos in code Code2Video-AA tests models across 114 design briefs, tasking them with turning each product or brand marketing brief into a @HyperFrames_ HTML composition, which we render to video. Model outputs are evaluated through pairwise comparisons with other model outputs across five axes: Engagement, Prompt Intent, Composition, Temporal Consistency, and Craft. We use the outcomes of these comparisons to calculate an Elo rating for each axis, which are then averaged to produce the headline Code2Video-AA Elo. Code2Video-AA was built in partnership with @HeyGen. Claude Opus 5.5 (max with fallback, 1536) leads Code2Video-AA by 93 points, but costs $3.08 per task, ~10x more than GPT-6.1 Sol (max, 1369), which trails by 167 points. Key findings: ➤ Claude Opus 5.5 leads on all five judge axes: Models are judged across 5 axes: Engagement, Prompt Intent, Composition, Temporal Consistency, and Craft. Claude Sonnet 5.5 (max with fallback, 1443) and GPT-6 Astra (max, 1427) are statistically tied for second in overall score ➤ Cost per task varies more than 300x, from $0.01 to $3.89: GPT-6.1 Sol (max), fourth, has a Cost per Task of $0.31, about 90% lower than Claude Opus 5.5 (max with fallback) at $3.08 ➤ Open weights models trail by at least 483 points: DeepSeek V4.1 Flash (max, 1053), MiMo-V2.6-Flash (1052) and MiMo-V2.6-Pro (1052) are the top open weights models, at $0.01 to $0.08 per task Code2Video-AA details: ➤ Models are tested on a single response without tools. We render it with HyperFrames, with no network access ➤ A judge (Gemini 4 Argon, high) compares pair of videos across five axes: engagement, prompt intent, composition, temporal consistency and craft, with the final score being calculated from a combination of each individual axis Elo ➤ Over 35,000 verdicts rank 27 models by Elo, with 95% confidence intervals More results, example videos and how Code2Video-AA works follow in the thread below

OpenAI has released GPT-6.1 Terra, pushing the cost efficiency frontier. It is the first Terra-class model since GPT-5.6, and is more intelligent for less than a third of the cost Pricing per token is half or less that of GPT-5.6 Terra - at $1/$5 per million input/output tokens vs $2/$12 for the previous model. Cache write has also halved, now at $1.25 per million tokens. Cached input is one quarter at $0.05 vs $0.20 per million tokens. Key takeaways: ➤ Pareto-optimal for Intelligence vs Cost at low and medium effort: At max effort, GPT-6.1 Terra gains 6 points in the Artificial Analysis Intelligence Index over GPT-5.6 Terra, at less than a third of the cost per task ($0.40 vs $1.40). At lower effort levels the gain is larger, between 7 and 13 points. The low and medium effort levels are Pareto-optimal for cost efficiency, while the other effort levels sit behind the frontier. At max effort, the model scores 4 points below GPT-6.1 Sol and 10 points above GPT-6 Luna in the Intelligence Index. ➤ Pushes the Coding Agent Index cost efficiency frontier: With a score of 48 in the newly released Artificial Analysis Coding Agent Index v2.0, the model is 6 points below GPT-6.1 Sol (54) and 17 points above GPT-6 Luna (31). The xhigh and low effort settings sit on the frontier of cost efficiency in the Coding Agent Index, while other effort levels are behind the frontier. ➤ 50 point drop in hallucination rate: GPT-6.1 Terra guesses 38% of the time when it does not know the answer to a question in AA-Omniscience, our private knowledge and hallucination benchmark. Previously, GPT-5.6 Terra's hallucination rate was 88%. This significant reduction in hallucination is coupled with a 2 point increase in AA-Omniscience accuracy - meaning the model simultaneously answers more questions correctly. ➤ 20 point jump in Terminal-Bench 4.0, small improvements across most evaluations: The model makes significant progress in the terminal-based agentic benchmark, Terminal-Bench 4.0, with a gain of ~20 points, now rivaling GPT-6.1 Sol at max effort. Gains across most other evaluations are smaller, for example, 5 points in GDP.pdf. Congratulations @OpenAI and @sama on the launch!

Kimi K3.1 Pro is among the leading models in agentic knowledge work and coding, but early signs suggest it will rank lower on AA-3D We are still collecting the preference votes for its Elo on AA-3D, our benchmark for how well AI models build interactive 3D scenes. In agentic knowledge work and coding, it is among the leading models on GDPval-AA and Terminal-Bench 4.0. Its predecessor, Kimi K3, is 12th of 22 models on AA-3D. Here is one of its AA-3D scenes: a 492 m mountain-bike descent with berms, roots, a rock garden, a drop, and a jump. A demo run rides it from top to bottom, and you can take over the steering, pedaling, and braking at any point. Congratulations to @Kimi_Moonshot on the release!

Kimi K3.1 Pro launches at 4th on the Artificial Analysis Intelligence Index scoring 55, making large gains in agentic performance; it will be the leading open-weights model once its weights are published @Kimi_Moonshot's new reasoning model gains 11 points over Kimi K3 (max, 44). It trails only Claude Opus 5.5 (max with fallback, 58), Claude Sonnet 5.5 (max with fallback, 56), and Muse Wonder Full (max, 56) on the Intelligence Index. Once the weights for K3.1 Pro are released it will be the highest-scoring open weights model and the gap between the leading open weights and proprietary models on the Intelligence Index will narrow from 12 points to 3. Key results: ➤ Large gains in agentic performance: Kimi K3.1 Pro scores 3rd on GDPval-AA with 1,846 Elo, up 312 Elo from Kimi K3, behind only Claude Opus 5.5 (1,866 Elo) and Muse Wonder Full (1,857 Elo). Kimi K3.1 Pro also achieves the highest score of any model on Terminal-Bench 4.0. K3.1 Pro scores 67%, up from 13% for Kimi K3 and ahead of Claude Sonnet 5.5 (64%). AutomationBench-AA, which tests agentic SaaS workflows, rises from 58% to 73%. ➤ Gains in other evals are more moderate: Humanity's Last Exam rises from 47% to 52%, CritPt from 23% to 25%, and SciCode from 59% to 61%. AA-Omniscience Index rises from 20 to 30, as the hallucination rate falls from 53% to 44% and accuracy rises from 48% to 51%. ➤ Cost per Task roughly equal to Kimi K3 at $1.98 per Task: K3.1 Pro uses 55K output tokens per Intelligence Index task, 14% more than Kimi K3 at 48K, while its first-party token prices are 33% lower. Additional model details: ➤ Context window: 1M tokens. ➤ Size: 2.8T total, 104B active parameters ➤ Input modalities: Text and image ➤ Weights and license: Weights currently not published, Moonshot is expected to release the weights soon after launch ➤ Pricing: $2/$10/$0.20 per 1M input/output/cache hit tokens, 33% lower than Kimi K3 on all three rates. Congratulations to the team at @Kimi_Moonshot on the release!

DeepSeek V4.1 Flash 1017 is the new #1# openly-available model in the Artificial Analysis Cyber Index. Other proprietary models may offer better cyber defense performance but are restricted access @deepseek_ai recently released DeepSeek V4.1 Flash 1017, an update to their flagship open-weights model DeepSeek V4.1 Flash. With an Artificial Analysis Cyber Index score of 57%, DeepSeek V4.1 Flash 1017 is the #1# openly-available model for cyber defense capabilities. The Artificial Analysis Cyber Index measures capability on agentic cyber defense tasks and is composed of three benchmarks: CWE-Bench-AA, DeepsecBench-AA and CyberGym-E2E-AA. The Cyber Index primarily evaluates models through their publicly accessible APIs, reflecting the cyber capabilities available to developers. Many models, particularly proprietary models from OpenAI, Anthropic and Google, may have stronger defensive cyber capabilities but encounter safety refusals that reduce their overall scores. GPT-6 Sol (Daybreak Blue, max), available through OpenAI’s trusted-access program, leads the Index, scoring 32 points above its public counterpart driven by fewer refusals. DeepSeek V4.1 Flash 1017 encounters no safety refusals. Key results: ➤ DeepSeek V4.1 Flash 1017 has the highest score among publicly-available models on the Artificial Analysis Cyber Index and #2# position overall. DeepSeek V4.1 Flash 1017 (max) scores 57% ahead of the previously leading publicly-available models Grok 4.7 (xhigh) and MiMo-V2.6-Pro by 1 point. It places #2# overall, with the trusted-access model GPT-6 Sol (Daybreak Blue, max) leading by 12 points with a score of 69% ➤ DeepSeek V4.1 Flash 1017 improves on its predecessor DeepSeek V4.1 Flash by 16 points. DeepSeek V4.1 Flash 1017 (max)’s 16-point improvement from 41% to 57% is driven by a substantial 53-point improvement on CyberGym-E2E-AA from 23% to 76%. On CWE-Bench-AA and DeepsecBench-AA it scores 69% and 25%, compared to 68% and 31% for DeepSeek V4.1 Flash (max) ➤ With a Cost per Task of $0.42, DeepSeek V4.1 Flash 1017 sits on the Capability vs. Cost Pareto frontier. DeepSeek V4.1 Flash 1017 costs $0.42 per task, more than 2x MiMo-V2.6-Pro (56%) at $0.18 per task but 4x cheaper than GPT-6 Sol (Daybreak Blue, max, 69%) at $1.77 per task