Gemma 4 31B
No shutdown before this date. Gemma 4 31B stays on Amazon Bedrock until at least Mar 31, 2027. Google names GLM-5.3-Flash as the replacement.
≥ 6 mo- Earliest retirement
- ≥ 168 daysMar 31, 2027 · Amazon Bedrock
- GPQA Diamond
- 75.8%#71 of 108 · Epoch AI
- OTIS Mock AIME 2024-2025
- 73.3%#66 of 105 · Epoch AI
- Context window
- 262.1Ktokens
- Released
- Apr 2, 2026Google
Dates on every platform
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| Gemini APIGoogle: Gemini API release notes (opens in a new tab) · checked Oct 14, 2026 | gemma-4-31b-itActive | – | None set | – |
| Amazon BedrockAmazon Web Services: Amazon Bedrock model card: Gemma 4 31B (opens in a new tab) · checked Oct 14, 2026 | google.gemma-4-31bActive | – | ≥ 6 moNot sooner than | – |
| Together AITogether AI: Changelog (opens in a new tab) · checked Oct 14, 2026 | google/gemma-4-31B-itShut down | – | Shut down | GLM-5.3-Flash |
Free tier only; Google lists no paid price.
Served on the bedrock-mantle endpoint only.
Together announced removal for 14 September 2026; the deprecation history records it on 15 September 2026.
What breaks when you move off Gemma 4 31B
Intelligence
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| GPQA Diamond | Epoch AI | 75.8% ± 3.1% | #71 of 108 | Aug 6, 2026 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 73.3% ± 6.7% | #66 of 105 | Aug 6, 2026 |
| SimpleQA Verified | Epoch AI | 10.4% ± 1.0% | #66 of 68 | Aug 27, 2026 |
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 |
|---|---|---|---|---|---|
Amazon Bedrockgoogle.gemma-4-31b | $0.14 | – | $0.40 | Oct 1, 2026Current · Standard tier on the bedrock-mantle endpoint. | Amazon Web Services: AWS Price List API: AmazonBedrock, us-east-1 (opens in a new tab) |
Change history
Shutdowngpt-oss-20b and Gemma 4 31B leave Together serverlessTogether AI1 break
Together removed openai/gpt-oss-20b (replacement Qwen3.5 9B) and google/gemma-4-31B-it (replacement GLM-5.3 Flash) from serverless. The removal was announced for 14 September 2026 and is recorded on 15 September 2026.
- Platform
- Together AI
- Models
- gpt-oss-20b, Gemma 4 31B
- Announced
- AvailabilityRequests to the removed serverless model ID fail. The model can still run on an on-demand dedicated endpoint, billed per minute.
Source: Together AI, Deprecations (opens in a new tab); Together AI, Changelog (opens in a new tab)