Change feed
Every pricing change and breaking deprecation we've detected, newest first.
-
Meta reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
Llama 4 Maverick: 10 benchmark claims (via web search)
Full history & current snapshot → -
OpenAI reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
GPT-5.6 Sol: 9 benchmark claims (via web search)
Full history & current snapshot → -
Moonshot AI reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
Kimi K3: 11 benchmark claims (via web search)
Full history & current snapshot → -
Zhipu AI (Z.ai) reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
GLM-5.2: 5 benchmark claims (via web search)
Full history & current snapshot → -
Alibaba reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
Qwen3.8-Max: 8 benchmark claims (via web search)
Full history & current snapshot → -
DeepSeek reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
DeepSeek-V4-Pro-Max: 4 benchmark claims (via web search)
Full history & current snapshot → -
xAI reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
Grok 4.6: 12 benchmark claims (via web search)
Full history & current snapshot → -
Google reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
Gemini 3.1 Pro: 11 benchmark claims (via web search)
Full history & current snapshot → -
Anthropic reported benchmarks updated
Tracking since Jun 25, 2026 · 9 snapshots on file
Claude Opus 5: 12 benchmark claims (via web search)
Full history & current snapshot → -
Tracking since Jun 25, 2026 · 17 snapshots on file
Arena Elo (text, overall) updated.
Full history & current snapshot →View raw diff +23 −23
- claude-fable-5 1494 - claude-opus-4-8-high 1462 - claude-opus-5-high 1505 - claude-opus-5-max 1503 - deepseek-v3.2-thinking 1420 - deepseek-v4-pro-high-20260813 1443 - gemini-3.1-pro-preview 1479 - gemini-3.5-flash-lite 1437 - gemini-3.5-flash-medium 1476 - gemini-3.6-flash-high 1477 - gemini-3.7-flash-high 1491 - glm-5.3-max 1474 - gpt-5.5-high 1470 - gpt-5.6-terra-xhigh 1445 - hy3 1442 - kimi-k3-max 1477 - mimo-v2.5-pro 1464 - minimax-m3 1434 - muse-spark 1473 - muse-spark-1.1 1479 - muse-spark-1.2 (xHigh) 1488 - qwen3.7-plus 1457 - qwen3.8-max 1483 + claude-fable-5 1495 + claude-opus-4-8-high 1461 + claude-opus-5-high 1504 + claude-opus-5-max 1504 + deepseek-v4-pro-high-20260813 1439 + gemini-3.1-pro-preview 1480 + gemini-3.5-flash-lite 1436 + gemini-3.5-flash-medium 1475 + gemini-3.6-flash-high 1476 + gemini-3.7-flash-high 1490 + glm-5.3-max 1477 + gpt-5.5-high 1471 + gpt-5.6-terra-xhigh 1446 + hy3 1441 + kimi-k3-max 1476 + mimo-v2.5-pro 1465 + minimax-m3 1435 + muse-spark 1474 + muse-spark-1.1 1478 + muse-spark-1.2 (xHigh) 1487 + qwen3.7-plus 1456 + qwen3.8-27b 1441 + qwen3.8-max 1482
-
NVIDIA Nemotron models changed
Tracking since Jun 25, 2026 · 18 snapshots on file
NVIDIA-Nemotron-Parse-2.0 license changed from 'other' to 'openmdw-1.1'
- License changed from 'other' to 'openmdw-1.1'
Full history & current snapshot →View raw diff +1 −1
- nvidia/NVIDIA-Nemotron-Parse-2.0 license:other 2026-06-30 + nvidia/NVIDIA-Nemotron-Parse-2.0 license:openmdw-1.1 2026-06-30
-
Tracking since Jun 25, 2026 · 40 snapshots on file
Anthropic rate limits documented by model class with separate ITPM/OTPM tracking and cache-aware token counting; no per-tier RPM/TPM tiers shown, only per-model limits
Full history & current snapshot →View raw diff +18 −230
- Claude Platform Docs - Messages - Managed Agents - Admin - * - Resources - Best practices - Models & pricing - CLI, SDKs, and libraries - Claude API skill - Release notes - API reference - English - Console Log in - - Search Ctrl K - Include beta APIs- [x] - Using the API - Features overview Beta headers Errors - Messages Create a Message Count tokens in a Message - Batches - Managed Agents - Agents - Environments - Sessions - Deployments - Deployment Runs - Vaults - Memory Stores - Dreams - Files Upload File List Files Download File Get File Metadata Delete File - Models List Models Get a Model - Skills Create Skill List Skills Get Skill Delete Skill - Versions - Tunnels Create Tunnel Get Tunnel List Tunnels Archive Tunnel Reveal Tunnel Token Rotate Tunnel Token - Certificates - User Profiles Create User Profile List User Profiles Get User Profile Update User Profile Create Enrollment URL - Webhooks - Admin - Organizations - Invites - Users - RBAC Groups - RBAC Roles - Workspaces - API Keys - External Keys - Usage Report - Cost Report - Analytics - Spend Limits - Rate Limits - Service Accounts - Federation Issuers - Federation Rules - MCP Tunnels - Compliance API - Activities - Organizations - Groups
-
Anthropic model deprecations changed
Tracking since Jun 25, 2026 · 29 snapshots on file
Model deprecation page restructured with navigation elements added and content reorganized, but core model status information preserved with same deprecation dates.
- Claude Mythos Preview deprecated in favor of Claude Mythos 5
- Model status table shows Active models with deprecation dates like June 9 2027, July 24 2027, May 28 2027
Full history & current snapshot →View raw diff +192 −34
- Overview - Migrating to replacements - Notifications - Auditing model usage - Best practices - Deprecation downsides and mitigations - Model status - | | Active | N/A | Not sooner than June 9, 2027 | - | | Active | N/A | Not sooner than July 24, 2027 | - | | Active | N/A | Not sooner than May 28, 2027 | - | | Active | N/A | Not sooner than April 16, 2027 | - | | Active | N/A | Not sooner than February 5, 2027 | - | | Active | N/A | Not sooner than November 24, 2026 | - | | Retired | June 5, 2026 | August 5, 2026 | - | | Retired | April 14, 2026 | June 15, 2026 | - | | Active | N/A | Not sooner than June 30, 2027 | - | | Active | N/A | Not sooner than February 17, 2027 | - | | Active | N/A | Not sooner than September 29, 2026 | - | | Retired | April 14, 2026 | June 15, 2026 | - | | Retired | October 28, 2025 | February 19, 2026 | - | | Active | N/A | Not sooner than October 15, 2026 | - | | Retired | December 19, 2025 | February 19, 2026 | - | | Retired | February 19, 2026 | April 20, 2026 | - Deprecation history - 2026-06-05: Claude Opus 4.1 model - 2026-04-14: Claude Sonnet 4 and Claude Opus 4 models - 2026-02-19: Claude Haiku 3 model - 2025-12-19: Claude Haiku 3.5 model - 2025-10-28: Claude Sonnet 3.7 model - 2025-08-13: Claude Sonnet 3.5 models - 2025-06-30: Claude Opus 3 model - 2025-01-21: Claude 2, Claude 2.1, and Claude Sonnet 3 models - 2024-09-04: Claude 1 and Instant models - API parameter deprecations + Claude Platform Docs + Messages + Managed Agents + Admin + * + Resources + Best practices + Models & pricing + CLI, SDKs, and libraries + Claude API skill + Release notes + API reference + English + Console Log in + + Search Ctrl K + Models + Models overview Model IDs and versioning Choosing a model Optimizing for cost and intelligence Introducing Claude Fable 5 and Claude Mythos 5 What's new in Claude Sonnet 5 What's new in Claude Opus 5 Upgrade between model versions Model deprecations Model cards System prompts Pricing + Console + Loading + Loading + Loading + Loading + Loading + Loading + Loading
-
Tracking since Jun 25, 2026 · 40 snapshots on file
No numeric changes to rate limits detected; page structure and documentation expanded with navigation elements and footers.
now50token200,050tokensFull history & current snapshot →View raw diff +231 −20
- About rate limits - Spend limits - Reaching your spend cap - Once you reach your tier's spend cap, API usage pauses until 00 - UTC on the first day of the next month, unless you request a higher limit sooner. While usage is paused, API requests return HTTP 429: - Setting your own spend limit - Navigate to the Billing page Go to Settings > Billing in the Claude Console. - Open the spend limit editor - In the Spend limits section, click Adjust limit (or Set limit if no limit is currently set). - Adjust your spend limit - Enter a new value. Your spend limit cannot exceed your current tier's cap. - Rate limits - Cache-aware ITPM - Message Batches API - Managed Agents - Files API - Fast mode rate limits - Monitoring your rate limits in the Console - Requesting higher limits - Setting lower limits for Workspaces + Claude Platform Docs + Messages + Managed Agents + Admin + * + Resources + Best practices + Models & pricing + CLI, SDKs, and libraries + Claude API skill + Release notes + API reference + English + Console Log in + + Search Ctrl K + Include beta APIs- [x] + Using the API + Features overview Beta headers Errors + Messages Create a Message Count tokens in a Message + Batches + Managed Agents + Agents + Environments + Sessions + Deployments + Deployment Runs + Vaults + Memory Stores + Dreams + Files Upload File List Files Download File Get File Metadata Delete File + Models List Models Get a Model + Skills Create Skill List Skills Get Skill Delete Skill + Versions + Tunnels Create Tunnel Get Tunnel List Tunnels Archive Tunnel Reveal Tunnel Token Rotate Tunnel Token + Certificates + User Profiles Create User Profile List User Profiles Get User Profile Update User Profile Create Enrollment URL + Webhooks + Admin + Organizations
-
Anthropic model deprecations changed
Tracking since Jun 25, 2026 · 29 snapshots on file
Documentation page content was removed showing model deprecation information and navigation structure.
- claude-opus-4-1-20250805 model retired June 5, 2026 with sunset August 5, 2026
- claude-opus-4-20250514 model retired April 14, 2026 with sunset June 15, 2026
- claude-3-7-sonnet-20250219 model retired October 28, 2025 with sunset February 19, 2026
Full history & current snapshot →View raw diff +35 −193
- Claude Platform Docs - Messages - Managed Agents - Admin - * - Resources - Best practices - Models & pricing - CLI, SDKs, and libraries - Claude API skill - Release notes - API reference - English - Console Log in - - Search Ctrl K - Models - Models overview Model IDs and versioning Choosing a model Optimizing for cost and intelligence Introducing Claude Fable 5 and Claude Mythos 5 What's new in Claude Sonnet 5 What's new in Claude Opus 5 Upgrade between model versions Model deprecations Model cards System prompts Pricing - Console - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Loading - Claude Platform Docs - - - Solutions - AI agents - Code modernization - Coding - Customer support - Financial services - Government - Higher education - K-12 teachers - Life sciences - Partners - Claude on AWS - Claude on Google Cloud - Learn - Blog - Courses - Use cases - Connectors - Customer stories - Engineering at Anthropic - Events - Powered by Claude
-
xAI (Grok) API pricing changed
Tracking since Jun 25, 2026 · 8 snapshots on file
Added notice that not all models support Batch API requests, with details available on individual model pages.
- Batch API is not available for every model
- Details on model support located on each model page
Full history & current snapshot →View raw diff +1 −0
+ Not every model accepts Batch API requests. See Details on each model page. -
Tracking since Jun 25, 2026 · 2 snapshots on file
Added deepseek-v4-flash-vision-exp model with 2500 concurrency limit; clarified per-user_id isolation limits for all three models.
Full history & current snapshot →View raw diff +3 −3
- deepseek-v4-prodeepseek-v4-flash - Concurrency Limit5002500 - For API users with increased concurrency quotas, we will limit the total concurrency under your account, and we will also impose concurrency limits on each user_id you pass (an empty id is treated as a special user_id). For each user_id, the concurrency limit for deepseek-v4-pro is 500, and for deepseek-v4-flash is 2500. If a user_id exceeds its limit, requests with that user_id under your account will receive an HTTP 429 error code + deepseek-v4-prodeepseek-v4-flashdeepseek-v4-flash-vision-exp + Concurrency Limit50025002500 + For API users with increased concurrency quotas, we will limit the total concurrency under your account, and we will also impose concurrency limits on each user_id you pass (an empty id is treated as a special user_id). For each user_id, the concurrency limit for deepseek-v4-pro is 500, for deepseek-v4-flash is 2500, and for deepseek-v4-flash-vision-exp is 2500. If a user_id exceeds its limit, requests with that user_id under your account will receive an HTTP 429 error code
-
Tracking since Jun 25, 2026 · 9 snapshots on file
Concurrency limit raised to 3; cache-based pricing tiers removed
rate_limitConcurrency limit increased from 2 to 3pricing_structureCache hit/miss pricing tiers removed; standard per-token pricing now usedFull history & current snapshot →View raw diff +5 −4
- 1M INPUT TOKENS (CACHE HIT) - 1M INPUT TOKENS (CACHE MISS) - Concurrency Limit(2) - (2) For more details on concurrency limits, please refer to Rate Limit & Isolation + 1M INPUT TOKENS + 1M INPUT TOKENS + Concurrency Limit(3) + (2) Images sent to deepseek-v4-flash-vision-exp are converted into tokens based on their dimensions and billed as input tokens together with your text tokens. See Vision: Token Usage for the conversion rule. + (3) For more details on concurrency limits, please refer to Rate Limit & Isolation.
-
Tracking since Jun 25, 2026 · 40 snapshots on file
Anthropic API rate limits defined by model class with separate ITPM/OTPM tracking and cache-aware token counting for most models
Full history & current snapshot →View raw diff +18 −230
- Claude Platform Docs - Messages - Managed Agents - Admin - * - Resources - Best practices - Models & pricing - CLI, SDKs, and libraries - Claude API skill - Release notes - API reference - English - Console Log in - - Search Ctrl K - Include beta APIs- [x] - Using the API - Features overview Beta headers Errors - Messages Create a Message Count tokens in a Message - Batches - Managed Agents - Agents - Environments - Sessions - Deployments - Deployment Runs - Vaults - Memory Stores - Dreams - Files Upload File List Files Download File Get File Metadata Delete File - Models List Models Get a Model - Skills Create Skill List Skills Get Skill Delete Skill - Versions - Tunnels Create Tunnel Get Tunnel List Tunnels Archive Tunnel Reveal Tunnel Token Rotate Tunnel Token - Certificates - User Profiles Create User Profile List User Profiles Get User Profile Update User Profile Create Enrollment URL - Webhooks - Admin - Organizations - Invites - Users - RBAC Groups - RBAC Roles - Workspaces - API Keys - External Keys - Usage Report - Cost Report - Analytics - Spend Limits - Rate Limits - Service Accounts - Federation Issuers - Federation Rules - MCP Tunnels - Compliance API - Activities - Organizations - Groups
-
Tracking since Jun 25, 2026 · 40 snapshots on file
Anthropic rate limits documented per usage tier with model-specific ITPM/OTPM, cache-aware token counting, and separate limits for Batches/Agents/Files APIs.
now50token200,050tokensFull history & current snapshot →View raw diff +232 −19
- About rate limits - Spend limits - Reaching your spend cap - Once you reach your tier's spend cap, API usage pauses until 00 UTC on the first day of the next month, unless you request a higher limit sooner. While usage is paused, API requests return HTTP 429: - Setting your own spend limit - Navigate to the Billing page Go to Settings > Billing in the Claude Console. - Open the spend limit editor - In the Spend limits section, click Adjust limit (or Set limit if no limit is currently set). - Adjust your spend limit - Enter a new value. Your spend limit cannot exceed your current tier's cap. - Rate limits - Cache-aware ITPM - Message Batches API - Managed Agents - Files API - Fast mode rate limits - Monitoring your rate limits in the Console - Requesting higher limits - Setting lower limits for Workspaces + Claude Platform Docs + Messages + Managed Agents + Admin + * + Resources + Best practices + Models & pricing + CLI, SDKs, and libraries + Claude API skill + Release notes + API reference + English + Console Log in + + Search Ctrl K + Include beta APIs- [x] + Using the API + Features overview Beta headers Errors + Messages Create a Message Count tokens in a Message + Batches + Managed Agents + Agents + Environments + Sessions + Deployments + Deployment Runs + Vaults + Memory Stores + Dreams + Files Upload File List Files Download File Get File Metadata Delete File + Models List Models Get a Model + Skills Create Skill List Skills Get Skill Delete Skill + Versions + Tunnels Create Tunnel Get Tunnel List Tunnels Archive Tunnel Reveal Tunnel Token Rotate Tunnel Token + Certificates + User Profiles Create User Profile List User Profiles Get User Profile Update User Profile Create Enrollment URL + Webhooks + Admin + Organizations + Invites
-
Tracking since Jun 25, 2026 · 19 snapshots on file
GitHub API: 5 additive changes
Full history & current snapshot →View raw diff +5 −0
+ New endpoint: GET /repos/{owner}/{repo}/stargazers/count + New optional parameter: before (query) + New optional parameter: after (query) + New optional parameter: before (query) + New optional parameter: after (query)
-
Tracking since Jun 25, 2026 · 2 snapshots on file
Pro plan pricing reduced with new annual billing model and explicit parallel agent limits introduced
Pro · price$25/month→$20/month, billed annuallyPro · price$100/month→$95/month, billed annuallyPro · rate_limitUnknown→2 parallel agents (lower tier) / 10 parallel agents (higher tier)Full history & current snapshot →View raw diff +12 −9
- FreeFor exploring what’s possible - Free daily Agent credits - $25$20per monthbilled annually - $25 of monthly credits - $100$95per monthbilled annually - $100 monthly credits - Join Replit Pro - CustomFor enterprise-grade security & controlsAll Pro features plus custom seat limits, SSO / SAML, advanced privacy controls, and more - Everything in Pro + FreeDaily agent credits + Free daily usage + $25 $20 / month, billed annually2 parallel agents + More free agent usage + $25 towards most powerful models + Plan mode + Pro + $100 $95 / month, billed annually10 parallel agents + Even more free agent usage + $100 towards most powerful models + Upgrade to Pro + CustomEnterprise-grade security and control
-
Tracking since Jun 25, 2026 · 4 snapshots on file
Removed six T5-Gemma model variants with gemma license dated 2025-06-19.
- google/t5gemma-l-l-prefixlm-it
- google/t5gemma-ml-ml-ul2
- google/t5gemma-xl-xl-ul2
Full history & current snapshot →View raw diff +0 −6
- google/t5gemma-l-l-prefixlm-it license:gemma 2025-06-19 - google/t5gemma-ml-ml-prefixlm license:gemma 2025-06-19 - google/t5gemma-ml-ml-prefixlm-it license:gemma 2025-06-19 - google/t5gemma-ml-ml-ul2 license:gemma 2025-06-19 - google/t5gemma-ml-ml-ul2-it license:gemma 2025-06-19 - google/t5gemma-xl-xl-ul2 license:gemma 2025-06-19
-
Tracking since Jun 25, 2026 · 40 snapshots on file
Anthropic enforces tiered rate limits by model with separate ITPM/OTPM tracking; cached input tokens exempt from ITPM calculation.
Full history & current snapshot →View raw diff +26 −214
- Claude Platform Docs - Messages - Managed Agents - Admin - * - Resources - Best practices - Models & pricing - CLI, SDKs, and libraries - Claude API skill - Release notes - API reference - English - Console Log in - - Search Ctrl K - Include beta APIs- [x] - Using the API - Features overview Beta headers Errors - Messages Create a Message Count tokens in a Message - Batches - Managed Agents - Agents - Environments - Sessions - Deployments - Deployment Runs - Vaults - Memory Stores - Dreams - Models List Models Get a Model - Files Upload File List Files Download File Get File Metadata Delete File - Skills Create Skill List Skills Get Skill Delete Skill - Versions - Tunnels Create Tunnel Get Tunnel List Tunnels Archive Tunnel Reveal Tunnel Token Rotate Tunnel Token - Certificates - User Profiles Create User Profile List User Profiles Get User Profile Update User Profile Create Enrollment URL - Webhooks - Admin - Organizations - Invites - Users - RBAC Groups - RBAC Roles - Workspaces - API Keys - External Keys - Usage Report - Cost Report - Analytics - Spend Limits - Rate Limits - Service Accounts - Federation Issuers - Federation Rules - MCP Tunnels - Compliance API - Activities - Organizations - Groups
-
Tracking since Jun 25, 2026 · 40 snapshots on file
Page structure normalized; substantive rate limits remain: Start/Build/Scale tiers with per-model ITPM/OTPM limits, spend caps, and cache-aware token counting unchanged.
now50token200,050tokensFull history & current snapshot →View raw diff +212 −16
- About rate limits - Spend limits - Navigate to the Billing page Go to Settings > Billing in the Claude Console. - Open the spend limit editor - In the Spend limits section, click Adjust limit (or Set limit if no limit is currently set). - Adjust your spend limit - Enter a new value. Your spend limit cannot exceed your current tier's cap. - Rate limits - Cache-aware ITPM - Message Batches API - Managed Agents - Files API - Fast mode rate limits - Monitoring your rate limits in the Console - Requesting higher limits - Setting lower limits for Workspaces + Claude Platform Docs + Messages + Managed Agents + Admin + * + Resources + Best practices + Models & pricing + CLI, SDKs, and libraries + Claude API skill + Release notes + API reference + English + Console Log in + + Search Ctrl K + Include beta APIs- [x] + Using the API + Features overview Beta headers Errors + Messages Create a Message Count tokens in a Message + Batches + Managed Agents + Agents + Environments + Sessions + Deployments + Deployment Runs + Vaults + Memory Stores + Dreams + Models List Models Get a Model + Files Upload File List Files Download File Get File Metadata Delete File + Skills Create Skill List Skills Get Skill Delete Skill + Versions + Tunnels Create Tunnel Get Tunnel List Tunnels Archive Tunnel Reveal Tunnel Token Rotate Tunnel Token + Certificates + User Profiles Create User Profile List User Profiles Get User Profile Update User Profile Create Enrollment URL + Webhooks + Admin + Organizations + Invites + Users + RBAC Groups + RBAC Roles
-
Tracking since Jun 25, 2026 · 17 snapshots on file
Arena Elo (text, overall) updated.
Full history & current snapshot →View raw diff +33 −33
- claude-fable-5 1493 - claude-opus-4-8-high 1461 - claude-opus-5-max 1508 - claude-sonnet-5-high 1443 - deepseek-v3.1-terminus-thinking 1420 - deepseek-v3.2-exp 1422 - deepseek-v4-flash 1431 - deepseek-v4-pro-high-preview 1446 - ernie-5.0-0110 1445 - ernie-5.0-preview-1203 1443 - gemini-3.1-pro-preview 1480 - gemini-3.5-flash-high 1482 - gemini-3.6-flash-high 1480 - glm-5.1 1463 - gpt-5.4 1452 - gpt-5.5-high 1471 - gpt-5.5-instant 1420 - gpt-5.6-luna-xhigh 1428 - gpt-5.6-sol-xhigh 1453 - gpt-5.6-terra-xhigh 1446 - grok-4.5 1451 - grok-4.6-high 1446 - hy3 1441 - inkling 1441 - kimi-k3-max 1476 - mimo-v2-pro 1436 - mimo-v2.5 1428 - mimo-v2.5-pro 1465 - minimax-m3 1435 - mistral-large-3 1429 - muse-spark-1.1 1477 - qwen3-235b-a22b-instruct-2507 1419 - qwen3.8-max 1492 + claude-fable-5 1494 + claude-opus-4-8-high 1462 + claude-opus-5-max 1503 + claude-sonnet-5-high 1442 + deepseek-v3.2-exp 1423 + deepseek-v4-flash 1432 + deepseek-v4-pro-high-20260813 1443 + deepseek-v4-pro-high-preview 1445 + ernie-5.0-0110 1444 + ernie-5.0-preview-1203 1442 + gemini-3.1-pro-preview 1479 + gemini-3.5-flash-high 1483 + gemini-3.6-flash-high 1477 + gemini-3.7-flash-high 1491 + glm-5.1 1464 + glm-5.3-max 1474 + gpt-5.4 1453 + gpt-5.5-high 1470 + gpt-5.6-luna-xhigh 1429 + gpt-5.6-sol-xhigh 1454 + gpt-5.6-terra-xhigh 1445 + grok-4.5 1452 + grok-4.6-high 1444 + hy3 1442 + inkling 1439 + kimi-k3-max 1477 + mimo-v2-pro 1437
-
GPQA Diamond (Epoch) scores changed
Tracking since Jun 25, 2026 · 18 snapshots on file
GPQA Diamond accuracy (0–1) updated.
Full history & current snapshot →View raw diff +1 −0
+ deepseek-v4-pro-0813_max 0.917 -
Tracking since Jun 25, 2026 · 40 snapshots on file
Anthropic publishes per-model RPM and separate ITPM/OTPM limits (cached tokens excluded); organization tier determines spend caps ($500–$200k monthly) and can request higher limits.
Full history & current snapshot →View raw diff +17 −209
- Claude Platform Docs - Messages - Managed Agents - Admin - * - Resources - Best practices - Models & pricing - CLI, SDKs, and libraries - Claude API skill - Release notes - API reference - English - Console Log in - - Search Ctrl K - Include beta APIs- [x] - Using the API - Features overview Beta headers Errors - Messages Create a Message Count tokens in a Message - Batches - Managed Agents - Agents - Environments - Sessions - Deployments - Deployment Runs - Vaults - Memory Stores - Dreams - Models List Models Get a Model - Files Upload File List Files Download File Get File Metadata Delete File - Skills Create Skill List Skills Get Skill Delete Skill - Versions - Tunnels Create Tunnel Get Tunnel List Tunnels Archive Tunnel Reveal Tunnel Token Rotate Tunnel Token - Certificates - User Profiles Create User Profile List User Profiles Get User Profile Update User Profile Create Enrollment URL - Webhooks - Admin - Organizations - Invites - Users - RBAC Groups - RBAC Roles - Workspaces - API Keys - External Keys - Usage Report - Cost Report - Analytics - Spend Limits - Rate Limits - Service Accounts - Federation Issuers - Federation Rules - MCP Tunnels - Compliance API - Activities - Organizations - Groups
-
deepseek-coder-1.3b-base: Epoch Capabilities Index (ECI) ↑ 62.02 → 62.28
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Cerebras-GPT-13B: Epoch Capabilities Index (ECI) ↑ 81.66 → 81.79
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
starcoder2-3b: Epoch Capabilities Index (ECI) ↑ 87.33 → 87.44
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
deepseek-coder-6.7b-base: Epoch Capabilities Index (ECI) ↑ 88.2 → 88.31
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
dolly-v2-12b: Epoch Capabilities Index (ECI) ↑ 88.25 → 88.36
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Baichuan-7B: Epoch Capabilities Index (ECI) ↑ 89.05 → 89.15
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
phi-1_5: Epoch Capabilities Index (ECI) ↑ 90.19 → 90.29
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
xgen-7b-8k-base: Epoch Capabilities Index (ECI) ↑ 92.09 → 92.19
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
starcoder2-7b: Epoch Capabilities Index (ECI) ↑ 92.24 → 92.33
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemma-2b: Epoch Capabilities Index (ECI) ↑ 92.91 → 93
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
mpt-7b: Epoch Capabilities Index (ECI) ↑ 93.31 → 93.4
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
falcon-7b: Epoch Capabilities Index (ECI) ↑ 93.83 → 93.92
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
deepseek-coder-33b-base: Epoch Capabilities Index (ECI) ↑ 95.08 → 95.16
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Baichuan-2-7B-Base: Epoch Capabilities Index (ECI) ↑ 95.1 → 95.17
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
LLaMA-7B: Epoch Capabilities Index (ECI) ↑ 95.47 → 95.56
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-2-7b: Epoch Capabilities Index (ECI) ↑ 97.91 → 97.98
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
LLaMA-13B: Epoch Capabilities Index (ECI) ↑ 99.47 → 99.54
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
mpt-30b: Epoch Capabilities Index (ECI) ↑ 99.55 → 99.62
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
mpt-30b-instruct: Epoch Capabilities Index (ECI) ↑ 99.55 → 99.62
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
INTELLECT-1-Instruct: Epoch Capabilities Index (ECI) ↑ 99.76 → 99.82
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Qwen2.5-Coder-1.5B: Epoch Capabilities Index (ECI) ↑ 101.95 → 102
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Baichuan-2-13B-Base: Epoch Capabilities Index (ECI) ↑ 102.16 → 102.22
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
falcon-40b-instruct: Epoch Capabilities Index (ECI) ↑ 103.46 → 103.51
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
falcon-40b: Epoch Capabilities Index (ECI) ↑ 103.46 → 103.51
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Yi-6B-Chat: Epoch Capabilities Index (ECI) ↑ 103.78 → 103.83
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Yi-6B: Epoch Capabilities Index (ECI) ↑ 103.78 → 103.83
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
starcoder2-15b: Epoch Capabilities Index (ECI) ↑ 104.01 → 104.06
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-2-34b: Epoch Capabilities Index (ECI) ↑ 104.23 → 104.28
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-2-13b-chat: Epoch Capabilities Index (ECI) ↑ 105.22 → 105.27
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-2-13b: Epoch Capabilities Index (ECI) ↑ 105.22 → 105.27
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Qwen-7B: Epoch Capabilities Index (ECI) ↑ 105.89 → 105.93
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
LLaMA-33B: Epoch Capabilities Index (ECI) ↑ 106.5 → 106.54
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Nemotron-4 15B: Epoch Capabilities Index (ECI) ↑ 106.82 → 106.86
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
phi-2: Epoch Capabilities Index (ECI) ↑ 107.02 → 107.06
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
falcon-11b: Epoch Capabilities Index (ECI) ↑ 108.7 → 108.73
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
LLaMA-65B: Epoch Capabilities Index (ECI) ↑ 109.34 → 109.37
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemma-7b: Epoch Capabilities Index (ECI) ↑ 111.21 → 111.24
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
falcon-180B: Epoch Capabilities Index (ECI) ↑ 111.38 → 111.4
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Qwen-14B-Chat: Epoch Capabilities Index (ECI) ↑ 112.27 → 112.3
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Qwen-14B: Epoch Capabilities Index (ECI) ↑ 112.27 → 112.3
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Qwen2.5-Coder-7B-Instruct: Epoch Capabilities Index (ECI) ↑ 112.39 → 112.41
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Qwen2.5-Coder-7B: Epoch Capabilities Index (ECI) ↑ 112.39 → 112.41
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gpt-3.5-turbo-0613: Epoch Capabilities Index (ECI) ↑ 112.61 → 112.63
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-2-70b-chat: Epoch Capabilities Index (ECI) ↑ 113.24 → 113.25
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-2-70b-hf: Epoch Capabilities Index (ECI) ↑ 113.24 → 113.25
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-2-70b-chat-hf: Epoch Capabilities Index (ECI) ↑ 113.24 → 113.25
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gpt-3.5-turbo-0125: Epoch Capabilities Index (ECI) ↓ 114.03 → 114.02
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Meta-Llama-3-8B: Epoch Capabilities Index (ECI) ↓ 116.03 → 116.02
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Meta-Llama-3-8B-Instruct: Epoch Capabilities Index (ECI) ↓ 116.03 → 116.02
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
StableBeluga2: Epoch Capabilities Index (ECI) ↑ 116.48 → 116.49
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemini-1.0-pro-001: Epoch Capabilities Index (ECI) ↓ 116.65 → 116.63
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
claude-3-haiku-20240307: Epoch Capabilities Index (ECI) ↓ 117.27 → 117.25
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Mixtral-8x7B-v0.1: Epoch Capabilities Index (ECI) ↓ 117.76 → 117.75
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
open-mixtral-8x7b: Epoch Capabilities Index (ECI) ↓ 117.76 → 117.75
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Mixtral-8x7B-Instruct-v0.1: Epoch Capabilities Index (ECI) ↓ 117.76 → 117.75
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gpt-3.5-turbo-1106: Epoch Capabilities Index (ECI) ↓ 117.93 → 117.92
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Mistral-Nemo-Instruct-2407: Epoch Capabilities Index (ECI) ↓ 118.03 → 118.02
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Mistral-Nemo-Base-2407: Epoch Capabilities Index (ECI) ↓ 118.03 → 118.02
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
open-mistral-nemo-2407: Epoch Capabilities Index (ECI) ↓ 118.03 → 118.02
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
claude-2.1: Epoch Capabilities Index (ECI) ↓ 118.05 → 118.03
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemma-2-9b: Epoch Capabilities Index (ECI) ↓ 119.26 → 119.24
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
claude-2.0: Epoch Capabilities Index (ECI) ↓ 119.25 → 119.24
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemma-2-9b-it: Epoch Capabilities Index (ECI) ↓ 119.26 → 119.24
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
claude-3-sonnet-20240229: Epoch Capabilities Index (ECI) ↓ 119.58 → 119.55
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
mistral-large-2402: Epoch Capabilities Index (ECI) ↓ 120.73 → 120.7
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Phi-3-medium-128k-instruct: Epoch Capabilities Index (ECI) ↓ 120.75 → 120.74
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
claude-instant-1.2: Epoch Capabilities Index (ECI) ↓ 120.8 → 120.79
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
claude-instant-1.1: Epoch Capabilities Index (ECI) ↓ 120.8 → 120.79
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Mixtral-8x22B-v0.1: Epoch Capabilities Index (ECI) ↓ 120.98 → 120.96
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Mixtral-8x22B-Instruct-v0.1: Epoch Capabilities Index (ECI) ↓ 120.98 → 120.96
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
open-mixtral-8x22b: Epoch Capabilities Index (ECI) ↓ 120.98 → 120.96
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Phi-3-small-8k-instruct: Epoch Capabilities Index (ECI) ↓ 121.33 → 121.32
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gpt-4-0613: Epoch Capabilities Index (ECI) ↓ 121.56 → 121.55
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemma-2-27b-it: Epoch Capabilities Index (ECI) ↓ 122.18 → 122.16
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemini-1.5-flash-0514: Epoch Capabilities Index (ECI) ↓ 122.26 → 122.24
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemini-1.5-flash-001: Epoch Capabilities Index (ECI) ↓ 122.26 → 122.24
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Meta-Llama-3-70B: Epoch Capabilities Index (ECI) ↓ 122.33 → 122.31
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Meta-Llama-3-70B-Instruct: Epoch Capabilities Index (ECI) ↓ 122.33 → 122.31
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
amazon.nova-pro-v1:0: Epoch Capabilities Index (ECI) ↑ 123.6 → 123.61
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
DeepSeek-V2: Epoch Capabilities Index (ECI) ↓ 124.34 → 124.33
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
qwen2-72b-instruct: Epoch Capabilities Index (ECI) ↓ 125 → 124.99
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-3.1-70B-Instruct: Epoch Capabilities Index (ECI) ↓ 125.26 → 125.25
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-3.2-90B-Vision-Instruct: Epoch Capabilities Index (ECI) ↓ 125.44 → 125.43
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gpt-4-32k-0314: Epoch Capabilities Index (ECI) ↓ 125.76 → 125.75
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gpt-4-0314: Epoch Capabilities Index (ECI) ↓ 125.76 → 125.75
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
claude-3-opus-20240229: Epoch Capabilities Index (ECI) ↓ 126.69 → 126.68
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
gemini-1.5-pro-001: Epoch Capabilities Index (ECI) ↓ 126.95 → 126.94
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
claude-3-5-haiku-20241022: Epoch Capabilities Index (ECI) ↓ 127.22 → 127.21
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
mistral-large-2407: Epoch Capabilities Index (ECI) ↓ 127.28 → 127.27
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
mistral-small-2503: Epoch Capabilities Index (ECI) ↓ 127.5 → 127.49
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
mistral-large-2411: Epoch Capabilities Index (ECI) ↓ 128.57 → 128.56
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot → -
Llama-3.1-405B: Epoch Capabilities Index (ECI) ↓ 128.76 → 128.75
Tracking since Jun 25, 2026 · 32 snapshots on file
Same model id, score moved on Epoch Capabilities Index — a silent re-evaluation or model swap.
Full history & current snapshot →
No changes match these filters.