Latest LLMs
38 new models in the last 90 days; the newest is Muse Spark Open.
Claude Opus 5.517.18%
Top in Text
Gemini 4 Argon1,523.9
Intelligence #1Claude Opus 5.557.6
Coding #1Claude Fable 5.181.6
Best value
MiMo-V2.6-Flash$0.28/1M out
Open weights #1
Kimi K3Arena 1,487.1
This week
Daily Agents records started on Oct 16, 2026; rank changes appear after 14 days.
Latest LLMs: full table
Top 30 of 92| # | Model | Agent | Text | AA index | Price $/1M tokens | Latency (s) & speed | Context | Released↓ | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Intel. | Coding | Input | Output | Lowest route | TTFT | Answer | tok/s | ||||||
| 1 |  Muse Spark OpenMetaOpen weights | – | – | 38.9 | – | $1.25 | $4.25 | $1.25 / $4.25 | 0.5 | 20 | 105 | 1.05M | Oct 15 |
| 2 | Muse Wonder PreliminaryMetaClosed | 6.42%Max Preliminary | 1508.9±10.6 | 52.4 | – | $4.00 | $20.00 | $4.00 / $20.00 | 33 | 45 | 174 | 1.05M | Oct 12 |
| 3 | Muse Wonder FullMeta | – | – | 55.8 | – | – | – | – | – | – | – | – | Oct 12 |
| 4 | Claude Haiku 5.5 HighAnthropic | 1.88%High | – | 37.8 | – | $0.10 | $0.50 | $0.10 / $0.50 | 20 | 20 | 171 | 1M | Oct 7 |
| 5 | Claude Haiku 5.5 LowAnthropic | 1.88%High | – | 29.5 | – | $0.10 | $0.50 | $0.10 / $0.50 | 6.8 | 6.8 | 158 | 1M | Oct 7 |
| 6 | Claude Haiku 5.5 MaxAnthropic | 1.88%High | – | 43.4 | – | $0.10 | $0.50 | $0.10 / $0.50 | 230 | 230 | 235 | 1M | Oct 7 |
| 7 | Claude Haiku 5.5 MediumAnthropic | 1.88%High | – | 34.5 | – | $0.10 | $0.50 | $0.10 / $0.50 | 10 | 10 | 169 | 1M | Oct 7 |
| 8 | Claude Haiku 5.5 XHighAnthropic | 1.88%High | – | 41.2 | – | $0.10 | $0.50 | $0.10 / $0.50 | 48 | 48 | 174 | 1M | Oct 7 |
| 9 |  Mistral Large 4 PreviewMistralClosed | -3.15% | 1431.8±7.3 | 38.4 | – | $1.36 | $4.18 | – | – | – | – | – | Oct 6 |
| 10 | ILing 3.1 FlashInclusionAI | – | – | 41.1 | – | $0.30 | $0.90 | $0.075 / $0.23 ⚠ | 1.1 | 10 | 217 | 262K | Oct 1 |
| 11 | Gemini 4 Argon PreliminaryGoogleClosed | 13.73%High | 1523.9±8.8 | 52.6 | – | $2.00 | $10.00 | $2.00 / $10.00 | – | – | – | 1.05M | Sep 30 |
| 12 |  GPT-6.1 Sol MaxOpenAIClosed | 14.78%Max | 1483.8±8.2 | 51.8 | – | $2.00 | $10.00 | $2.00 / $10.00 | 164 | 164 | 59 | 1.05M | Sep 29 |
| 13 | GPT-6.1 Sol XHighOpenAIClosed | 14.78%Max | 1483.8±8.2 | 51.0 | – | $2.00 | $10.00 | $2.00 / $10.00 | 110 | 110 | 52 | 1.05M | Sep 29 |
| 14 | GPT-6.1 Sol HighOpenAIClosed | 14.78%Max | 1483.8±8.2 | 50.2 | – | $2.00 | $10.00 | $2.00 / $10.00 | 36 | 36 | 52 | 1.05M | Sep 29 |
| 15 | GPT-6.1 Sol MediumOpenAIClosed | 14.78%Max | 1483.8±8.2 | 47.8 | – | $2.00 | $10.00 | $2.00 / $10.00 | 5.2 | 5.2 | 50 | 1.05M | Sep 29 |
| 16 | GPT-6.1 Sol LowOpenAIClosed | 14.78%Max | 1483.8±8.2 | 42.1 | – | $2.00 | $10.00 | $2.00 / $10.00 | 2.4 | 2.4 | 52 | 1.05M | Sep 29 |
| 17 | Claude Sonnet 5.5 MaxAnthropicClosed | 12.55%Max Preliminary | 1474.9±7.8 | 56.0 | – | $2.00 | $10.00 | $2.00 / $10.00 | 301 | 301 | 142 | 1M | Sep 28 |
| 18 | Claude Sonnet 5.5 XHighAnthropicClosed | 12.55%Max Preliminary | 1474.9±7.8 | 51.9 | – | $2.00 | $10.00 | $2.00 / $10.00 | 20 | 20 | 106 | 1M | Sep 28 |
| 19 | Claude Sonnet 5.5 HighAnthropicClosed | 12.55%Max Preliminary | 1474.9±7.8 | 46.8 | – | $2.00 | $10.00 | $2.00 / $10.00 | 9.9 | 9.9 | 102 | 1M | Sep 28 |
| 20 | Claude Sonnet 5.5 MediumAnthropicClosed | 12.55%Max Preliminary | 1474.9±7.8 | 40.8 | – | $2.00 | $10.00 | $2.00 / $10.00 | 1.3 | 1.3 | 98 | 1M | Sep 28 |
| 21 | Claude Sonnet 5.5 LowAnthropicClosed | 12.55%Max Preliminary | 1474.9±7.8 | 35.9 | – | $2.00 | $10.00 | $2.00 / $10.00 | 0.7 | 0.7 | 95 | 1M | Sep 28 |
| 22 | Claude Opus 5.5 MaxAnthropicClosed | 17.18%Max Preliminary | 1508.2±7.4 | 57.6 | – | $4.00 | $20.00 | $4.00 / $20.00 | 450 | 450 | 98 | 1M | Sep 22 |
| 23 | Claude Opus 5.5 XHighAnthropicClosed | 17.18%Max Preliminary | 1508.2±7.4 | 56.0 | – | $4.00 | $20.00 | $4.00 / $20.00 | 96 | 96 | 81 | 1M | Sep 22 |
| 24 | Claude Opus 5.5 HighAnthropicClosed | 17.18%Max Preliminary | 1508.2±7.4 | 53.6 | – | $4.00 | $20.00 | $4.00 / $20.00 | 21 | 21 | 76 | 1M | Sep 22 |
| 25 | Claude Opus 5.5 MediumAnthropicClosed | 17.18%Max Preliminary | 1508.2±7.4 | 51.2 | – | $4.00 | $20.00 | $4.00 / $20.00 | 15 | 15 | 76 | 1M | Sep 22 |
| 26 | Claude Opus 5.5 LowAnthropicClosed | 17.18%Max Preliminary | 1508.2±7.4 | 42.3 | – | $4.00 | $20.00 | $4.00 / $20.00 | 4.9 | 4.9 | 77 | 1M | Sep 22 |
| 27 | GPT-6 Sol MaxOpenAIClosed | 11.20%Max Preliminary | 1459.0±6.4 | 47.6 | – | $2.00 | $10.00 | $2.00 / $10.00 | – | – | – | 1.05M | Sep 22 |
| 28 | GPT-6 Sol XHighOpenAIClosed | 11.20%Max Preliminary | 1459.0±6.4 | 44.2 | – | $2.00 | $10.00 | $2.00 / $10.00 | – | – | – | 1.05M | Sep 22 |
| 29 | GPT-6 Sol HighOpenAIClosed | 11.20%Max Preliminary | 1459.0±6.4 | 42.4 | – | $2.00 | $10.00 | $2.00 / $10.00 | – | – | – | 1.05M | Sep 22 |
| 30 | GPT-6 Sol MediumOpenAIClosed | 11.20%Max Preliminary | 1459.0±6.4 | 39.8 | – | $2.00 | $10.00 | $2.00 / $10.00 | – | – | – | 1.05M | Sep 22 |
Arena scores are LMArena's Bradley-Terry scale (not Elo). Input/output prices are official API list prices per million tokens; the lowest route price is the cheapest OpenRouter quote. In this view the median Arena margin is ±6.3 while neighbouring models sit a median 3.1 apart (27 of 27 adjacent pairs overlap): gaps under 6.3 say nothing about which model ranks higher.
Expanded by tier: each reasoning-effort setting of a model gets its own row, labelled with Artificial Analysis’s own effort names (Max, XHigh, High, Medium, Low, …), and the colour bar on the left ties the rows of one model together. The score is that setting’s measurement; Arena scores and prices are model-level, so rows of the same model share them.
Not yet ranked by Arena 2
Tracked models that have no LMArena or Artificial Analysis score yet, newest first. They move into the table once scored.
On the Agents board, not yet in our catalogue 5
These models have an arena.ai Agents result but no model page here yet (no price data), so they are not ranked in the table above.
- THy4 previewtencent · 2.90% · arena.ai #21
- NNemotron 3 Ultranvidia · -8.29% · arena.ai #50
- TInklingthinky · -13.61% · arena.ai #52
- TInkling Smallthinky · -14.72% · arena.ai #53
- USolar Pro 4upstage · -20.18% · arena.ai #54
All 86 model pages
OpenAI 14
Anthropic 12
Google 9
Alibaba 8
Meta 7
DeepSeek 6
Moonshot AI 4
xAI 4
Xiaomi 4
Z.AI 4
Mistral 3
Z AI 2
ByteDance 1
CChina Mobile 1
IInclusionAI 1
IInstitute of Foundation Models 1
MiniMax 1
MMotif Technologies 1
SSapiens AI 1
SStepFun 1
Tencent 1
Sources & methodology
38 tracked models were released in the last 90 days; the newest is Muse Spark Open (Oct 15, 2026). The table is sorted by release date (from Artificial Analysis, falling back to the OpenRouter listing date).
- AGarena.ai Agents
Agents leaderboard: task success rate, 95% interval, sessions and rank (arena.ai Agents, via the public Hugging Face dataset lmarena-ai/leaderboard-dataset, CC BY 4.0). Hugging Face
- LMLMArena
Arena score, 95% confidence interval, votes and rank (overall Text leaderboard with style control, the same view arena.ai shows by default; dataset CC BY 4.0).
- AAArtificial Analysis
Intelligence Index, Coding Index, output speed, TTFT / first-answer latency, release dates and official prices (shown on this page only, not in the public API).
- OROpenRouter
Per-provider prices (the cheapest becomes “lowest route”), context length and listing dates.
- $Vendor pricing pages
“Input / Output” are official API list prices (USD per 1M tokens); when no official price exists we use the OpenRouter price and mark it with *.
How we compute it
- Ranking: by default, the arena.ai Agents task success rate (real users’ coding/tool agent tasks). When a model has several reasoning tiers (Max/High/xHigh), the best-scoring tier is used and labelled; fewer than 10,000 sessions or a 95% half-width above 3 points is marked Preliminary. Models without an agent result follow, ordered by LMArena Text (style control); the Text column is still sortable.
- Rank: by default, LMArena Text Arena score with style control (the same view arena.ai shows by default; Bradley-Terry, not Elo). The ± is the half-width of the 95% confidence interval, and neighbouring ranks often overlap. Models with fewer than 5,000 votes or ± above 10 are marked Preliminary. We do not invent a composite score.
- Intelligence / Coding Index: Artificial Analysis aggregates of public benchmarks. When a model has several reasoning-effort tiers, the table shows the best tier for the current sort column; switch to “by tier” to see each one.
- Prices are USD per 1M tokens. “Lowest route” is the cheapest OpenRouter provider’s input / output price; ⚠ marks routes more than 60% below list (usually promos or a single provider). The best-value board only includes models with Arena ≥ 1400 or Intelligence Index ≥ 40, sorted by official output price.
- Latency and speed: TTFT is the time to the first token (including thinking tokens); “Answer” is the time to the first answer token (includes the thinking phase for reasoning models); speed is output tokens per second. All are Artificial Analysis medians.
- Updates: Text and other data sync daily at 14:20 (UTC+8), the Agents board at 14:25. Agent rank changes appear only after 14 days of daily records; until then no arrows are shown.
- Updates: synced automatically every day at 14:20 (UTC+8). LMArena publishes on its own schedule, so the “Arena published” date can be older than the update date. “This week” compares against the daily archive from 7 days earlier.
- Open data: /api/public/llm-leaderboard.php (JSON, LMArena and OpenRouter fields only).
FAQ
What is the best AI model right now?
As of Oct 16, 2026, the arena.ai Agents leaderboard (success rate on real coding/tool tasks) is led by Claude Opus 5.5 (17.18%, 95% interval 13.80%–20.56%), ahead of GPT-6.1 Sol (14.78%). The top intervals overlap heavily, so the exact order is not reliable. On the LMArena Text leaderboard, Gemini 4 Argon leads (1,523.9).
Which good AI model is the cheapest?
Among models above the quality floor (Arena ≥ 1400 or Intelligence Index ≥ 40), MiMo-V2.6-Flash has the lowest official price: $0.14 input / $0.28 output per 1M tokens, followed by MiMo V2.5 ($0.28 output). See the best-value leaderboard.
What is the best open-source (open-weight) AI model?
By Arena score, the top open-weight model is Kimi K3 (Moonshot AI, 1,487.1), followed by MiMo-V2.6-Pro (1,479.2). We track 21 open-weight models with an Arena score — see the open-weight leaderboard. Open weights do not mean unrestricted commercial use; check each licence.
Which AI model is best for coding?
On the Artificial Analysis Coding Index, Claude Fable 5.1 is highest (81.6), followed by Claude Opus 5 (78.0). The index aggregates coding benchmarks and differs from the Arena score; see the full coding leaderboard.
Which AI model is the fastest?
The fastest measured output speed belongs to K2 Horizon 375B A23B (1021 tokens/s); the lowest time to first answer token belongs to K2 Horizon 375B A23B (3.65 s). These are Artificial Analysis medians; for reasoning models the first-answer latency includes thinking time.
Which AI models were released recently?
5 tracked models were released in the last 14 days, including Muse Spark Open (Oct 15, 2026), Muse Wonder (Oct 12, 2026), Muse Wonder Full (Oct 12, 2026), Claude Haiku 5.5 (Oct 7, 2026). New models usually take a few days to appear on LMArena; until then they are listed under “Not yet ranked on Arena”.
How often is the data updated, and where does it come from?
Synced automatically every day: the Agents board comes from arena.ai (public Hugging Face dataset, current release Oct 14, 2026), Text scores from LMArena (released Oct 15, 2026); intelligence and coding indices, speed and latency from Artificial Analysis; route prices and context from OpenRouter. This page was updated Oct 16, 2026.
Why do prices differ from the official site?
Input/output prices are official list prices in USD per million tokens. The lowest route price is the cheapest OpenRouter route, which may be a promotion or a single provider and can differ from local resellers.