Llama 3.3 70B Instruct
Shutting down. Llama 3.3 70B Instruct shuts down on Google Vertex AI on Oct 21, 2026. Meta names Muse Glimmer 30B as the replacement.
7 days- Next shutdown
- 7 daysOct 21, 2026 · Google Vertex AI
- Arena Text
- 1,318#108 of 128 · Arena
- GPQA Diamond
- 47.4%#94 of 108 · Epoch AI
- List price · $ per 1M
- $0.71 in $0.71 outAzure AI Foundry
- Context window
- 131.1Ktokens
Dates on every platform
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| Azure AI FoundryMicrosoft: Model retirement schedule - Microsoft Foundry (opens in a new tab) · checked Oct 14, 2026 | Llama-3.3-70B-InstructActive | – | None set | – |
| Amazon BedrockAmazon Web Services: Amazon Bedrock ListFoundationModels, us-east-1 (opens in a new tab) · checked Oct 14, 2026us-east-1, us-east-2, us-west-2 | meta.llama3-3-70b-instruct-v1:0Deprecated since Sep 30, 2026 · also us.meta.llama3-3-70b-instruct-v1:0 | – | 5 mo | – |
| Google Vertex AIGoogle Cloud: Open model deprecations (opens in a new tab) · checked Oct 14, 2026 | llama-3.3-70b-instruct-maasDeprecated | 7 days | – | |
| Together AITogether AI: Changelog (opens in a new tab) · checked Oct 14, 2026 | meta-llama/Llama-3.3-70B-Instruct-TurboDeprecated | – | 8 days | Muse Glimmer 30B |
| Fireworks AIFireworks AI: Llama 3.3 70B Instruct model page (opens in a new tab) · checked Oct 14, 2026 | accounts/fireworks/models/llama-v3p3-70b-instructShut down | – | Shut down | gpt-oss-120b |
New customers can't use Legacy models on Bedrock, and accounts that stop calling one for 15 days may lose access before end of life.
On 14 October 2026 ListFoundationModels still returned this model as active with no end-of-life date; the Bedrock lifecycle page lists the Legacy and end-of-life dates.
Bedrock lists the public extended access start date as to be announced.
In us-east-1, requests go through a cross-Region inference profile (listed in aliases).
Google suggests self-deploying the model from Model Garden or moving to another managed endpoint.
Supported by on-demand dedicated endpoints after removal from serverless.
Removed from serverless only; dedicated deployments of this model are unaffected.
What breaks when you move off Llama 3.3 70B Instruct
- Availability
New customers can't use Legacy models, and accounts that stop calling one for 15 days may lose access before the end-of-life date.
Gemma 3 4B IT, Gemma 3 12B IT, Gemma 3 27B IT, Llama 3 8B Instruct, Llama 3 70B Instruct, Llama 3.1 8B Instruct, Llama 3.1 70B Instruct, Llama 3.3 70B Instruct, Llama 4 Scout 17B 16E Instruct, Llama 4 Maverick 17B 128E Instruct, DeepSeek-R1, DeepSeek-V3.1, Qwen3-Coder-30B-A3B-Instruct, Qwen3-Coder-480B-A35B-Instruct, MiniMax-M2, MiniMax-M2.1, Devstral 2, Mistral 7B Instruct, Mixtral 8x7B Instruct, Voxtral Mini, Voxtral Small, Kimi K2 Thinking, NVIDIA Nemotron Nano 9B v2, NVIDIA Nemotron Nano 12B v2 VL, GLM-4.7, GLM-4.7-Flash · Amazon Bedrock ·
- Availability
Requests to the model ID fail after the end-of-life date. Bedrock does not migrate applications to another model.
Gemma 3 4B IT, Gemma 3 12B IT, Gemma 3 27B IT, Llama 3 8B Instruct, Llama 3 70B Instruct, Llama 3.1 8B Instruct, Llama 3.1 70B Instruct, Llama 3.3 70B Instruct, Llama 4 Scout 17B 16E Instruct, Llama 4 Maverick 17B 128E Instruct, DeepSeek-R1, DeepSeek-V3.1, Qwen3-Coder-30B-A3B-Instruct, Qwen3-Coder-480B-A35B-Instruct, MiniMax-M2, MiniMax-M2.1, Devstral 2, Mistral 7B Instruct, Mixtral 8x7B Instruct, Voxtral Mini, Voxtral Small, Kimi K2 Thinking, NVIDIA Nemotron Nano 9B v2, NVIDIA Nemotron Nano 12B v2 VL, GLM-4.7, GLM-4.7-Flash · Amazon Bedrock ·
- Tooling
You can't create new Provisioned Throughput or fine-tuning jobs for a Legacy model.
Gemma 3 4B IT, Gemma 3 12B IT, Gemma 3 27B IT, Llama 3 8B Instruct, Llama 3 70B Instruct, Llama 3.1 8B Instruct, Llama 3.1 70B Instruct, Llama 3.3 70B Instruct, Llama 4 Scout 17B 16E Instruct, Llama 4 Maverick 17B 128E Instruct, DeepSeek-R1, DeepSeek-V3.1, Qwen3-Coder-30B-A3B-Instruct, Qwen3-Coder-480B-A35B-Instruct, MiniMax-M2, MiniMax-M2.1, Devstral 2, Mistral 7B Instruct, Mixtral 8x7B Instruct, Voxtral Mini, Voxtral Small, Kimi K2 Thinking, NVIDIA Nemotron Nano 9B v2, NVIDIA Nemotron Nano 12B v2 VL, GLM-4.7, GLM-4.7-Flash · Amazon Bedrock ·
- Availability
- Availability
API requests to the retired model IDs fail after the retirement date.
DeepSeek-R1, DeepSeek-V3.2, DeepSeek-V3.1, GLM-5, GLM-4.7, gpt-oss-20b, Kimi K2 Thinking, Llama 3.3 70B Instruct, MiniMax-M2, Qwen3-235B-A22B-Instruct-2507, Qwen3-Coder-480B-A35B-Instruct, Qwen3-Next-80B-A3B-Instruct, Qwen3 Next 80B A3B Thinking · Google Vertex AI ·
- Availability
Intelligence
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| Arena Text | Arena | 1,318 ± 2 | #108 of 128 | Oct 8, 2026 |
| GPQA Diamond | Epoch AI | 47.4% ± 2.9% | #94 of 108 | Jan 27, 2025 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 5.1% ± 2.4% | #97 of 105 | Feb 25, 2025 |
Arena (arena.ai), https://arena.ai/leaderboard
Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource].
List prices
| Platform | Input | Cached | Output | In effect | Source |
|---|---|---|---|---|---|
Azure AI FoundryLlama-3.3-70B-Instruct | $0.71 | – | $0.71 | Jan 1, 2026Current · Global Standard rate from the US East 2 price list. | Microsoft: Azure Retail Prices API (Foundry Models, US East 2) (opens in a new tab) |
Amazon Bedrockmeta.llama3-3-70b-instruct-v1:0 | $0.72 | – | $0.72 | Oct 1, 2026Current · Standard on-demand tier in us-east-1. | Amazon Web Services: AWS Price List API: AmazonBedrock, us-east-1 (opens in a new tab) |
Together AImeta-llama/Llama-3.3-70B-Instruct-Turbo | $1.04 | – | $1.04 | May 29, 2026Current | Together AI: Changelog (opens in a new tab) |
Change history
Shutdown26 open-weight models reach end of life on Amazon BedrockAmazon Bedrock3 breaks
Bedrock moved 26 open-weight models from DeepSeek, Google, Meta, MiniMax, Mistral AI, Moonshot AI, NVIDIA, Qwen and Z.ai to Legacy on 30 September 2026. Requests to them fail after 30 March 2027. On 14 October 2026 ListFoundationModels still reported them as active.
- Platform
- Amazon Bedrock
- Models
- Gemma 3 4B IT, Gemma 3 12B IT, Gemma 3 27B IT, Llama 3 8B Instruct, Llama 3 70B Instruct, Llama 3.1 8B Instruct, Llama 3.1 70B Instruct, Llama 3.3 70B Instruct, Llama 4 Scout 17B 16E Instruct, Llama 4 Maverick 17B 128E Instruct, DeepSeek-R1, DeepSeek-V3.1, Qwen3-Coder-30B-A3B-Instruct, Qwen3-Coder-480B-A35B-Instruct, MiniMax-M2, MiniMax-M2.1, Devstral 2, Mistral 7B Instruct, Mixtral 8x7B Instruct, Voxtral Mini, Voxtral Small, Kimi K2 Thinking, NVIDIA Nemotron Nano 9B v2, NVIDIA Nemotron Nano 12B v2 VL, GLM-4.7, GLM-4.7-Flash
- Announced
- AvailabilityNew customers can't use Legacy models, and accounts that stop calling one for 15 days may lose access before the end-of-life date.
- AvailabilityRequests to the model ID fail after the end-of-life date. Bedrock does not migrate applications to another model.
- ToolingYou can't create new Provisioned Throughput or fine-tuning jobs for a Legacy model.
Source: Amazon Web Services, Amazon Bedrock User Guide: Model lifecycle (Legacy) (opens in a new tab); Amazon Web Services, Amazon Bedrock ListFoundationModels, us-east-1 (opens in a new tab)
ShutdownLlama 3.3 70B Instruct Turbo leaves Together serverlessTogether AI1 break
Together removes meta-llama/Llama-3.3-70B-Instruct-Turbo from serverless inference and recommends Muse Glimmer 30B.
- Platform
- Together AI
- Replacement
- Muse Glimmer 30B
- AvailabilityRequests to the removed serverless model ID fail. The model can still run on an on-demand dedicated endpoint, billed per minute.
ShutdownDeepSeek, GLM, Kimi, Llama, MiniMax, Qwen and GPT OSS endpoints retire on Agent PlatformGoogle Vertex AI1 break
Google retires its managed endpoints for DeepSeek, GLM, GPT OSS 20B, Kimi K2 Thinking, Llama 3.3 70B, MiniMax M2 and Qwen3 models on Gemini Enterprise Agent Platform. Google suggests self-deploying them from Model Garden.
- Platform
- Google Vertex AI
- Models
- DeepSeek-R1, DeepSeek-V3.2, DeepSeek-V3.1, GLM-5, GLM-4.7, gpt-oss-20b, Kimi K2 Thinking, Llama 3.3 70B Instruct, MiniMax-M2, Qwen3-235B-A22B-Instruct-2507, Qwen3-Coder-480B-A35B-Instruct, Qwen3-Next-80B-A3B-Instruct, Qwen3 Next 80B A3B Thinking
- Announced
- AvailabilityAPI requests to the retired model IDs fail after the retirement date.
Source: Google Cloud, Gemini Enterprise Agent Platform release notes (opens in a new tab); Google Cloud, Open model deprecations (opens in a new tab)
ShutdownFireworks removes eight older models from serverlessFireworks AI1 break
DeepSeek V3.1, DeepSeek V3.2, GLM 4.7, GLM 5, Qwen3 8B, both Qwen3 VL 30B A3B models and Llama 3.3 70B Instruct left serverless. Fireworks named Kimi K2.6, GLM 5.1, GPT-OSS 20B and GPT-OSS 120B as migration targets.
- Platform
- Fireworks AI
- AvailabilityServerless requests to these model IDs stop working after the date. Dedicated deployments are unaffected.