Available models
Posit Assistant lets you choose which large language model handles your request. Each model offers a different balance of capability, speed, and credit usage, and you select the active model in the Posit Assistant interface.
The following models may be available with your Posit AI Pass subscription. We add, update, and retire models as the field moves, so treat the model picker in Posit Assistant as the authoritative list.
Different models consume your Posit AI Pass usage credits at different rates. The table below expresses rates relative to Claude Sonnet 5.5, which serves as the baseline. For example, a request that uses 1% of your credits on Sonnet 5.5 uses roughly 0.5% on Haiku 4.5 and roughly 2% on Opus 5.5.
The Relative Usage Amount column shows the inverse: how much usage you get for the same credit spend. For example, GLM 5.3 Flash gives you roughly 10x as much usage as Sonnet 5.5 for the same credits.
| Model | Developer | Relative Credit Cost | Relative Usage Amount |
|---|---|---|---|
| Claude Haiku 5.5 | Anthropic | 0.06–0.3x | 3x–17x |
| Claude Sonnet 5.5 | Anthropic | 1x (baseline) | 1x (baseline) |
| Claude Opus 5.5 | Anthropic | 2x | 0.5x |
Note: Haiku 5.5 prompts over 100K tokens cost 5x as much per token.
Additional models are available under More models in the model picker:
| Model | Developer | Relative Credit Cost | Relative Usage Amount |
|---|---|---|---|
| GLM 5.3 Flash | Z.ai (open weights) | 0.1x | 10x |
| DeepSeek V4.1 Flash | DeepSeek (open weights) | 0.1x | 10x |
| GLM 5.3 | Z.ai (open weights) | 0.7x | 1.4x |
| Kimi K3.1 Pro | Moonshot AI (open weights) | 0.8x | 1.2x |
| Kimi K3 | Moonshot AI (open weights) | 1.2x | 0.8x |
Open weights models
Posit serves the open weights models through our agreement with Baseten, on GPU infrastructure that does not retain data. Posit does not store your conversation histories unless you opted in to data retention at sign-up. See your data preferences and the FAQ on Zero Data Retention.
The GLM models and Kimi K3 share a few characteristics worth knowing about:
- Speed: they stream almost twice as many tokens per second as the Anthropic models, which gives them a qualitatively different feel in interactive use.
- Tokenization: these models tokenize text more efficiently, so the real cost savings are larger than the per-token difference alone implies.
- Context: all have context windows of 250K tokens or more, and Posit AI Pass does not charge separately for cache writes on them.
For more on how these models fit into Posit AI Pass, see our post on Kimi K3 and GLM 5.2.
GLM 5.3
GLM 5.3 is available as a preview at roughly 0.7x the credit usage of Sonnet 5.5. Expect it to be a stronger fit for agentic coding than for visual or exploratory data analysis work, since it cannot see plots. Since it is a preview release, expect rougher edges and treat its output with a bit more scrutiny than the stable models.
GLM 5.3 Flash
GLM 5.3 Flash is a preview release and one of the two lowest-consumption models in Posit AI Pass, at roughly a tenth of the credit usage of Sonnet 5.5. Use it for quick, low-stakes tasks where speed and credit savings matter more than top-end capability, and review its output closely.
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash matches GLM 5.3 Flash as the lowest-consumption option, at roughly a tenth of the credit usage of Sonnet 5.5. Like GLM 5.3 Flash, it suits quick, low-stakes tasks where speed and credit savings matter more than top-end capability.
Kimi K3
Kimi K3 is the most capable open weights model in Posit AI Pass, and notably well-rounded compared to other open weights releases rather than strong in one narrow area. It draws credits at roughly 1.2x the rate of Sonnet 5.5.
Use it when you want a strong, fast open weights model for agentic work.
Claude models
Posit AI Pass sends Claude requests to Anthropic under a Zero Data Retention agreement.
Claude Haiku 5.5
A fast, lightweight model. Good for quick questions, simple edits, short refactors, and follow-ups that do not need deep reasoning. It draws credits at roughly 0.06x the rate of Sonnet 5.5, making it the Claude option that draws credits the slowest. Prompts over 100K tokens cost 5x as much per token, bringing the rate to roughly 0.3x, so it is most economical in shorter conversations.
Claude Sonnet 5.5
The recommended default for most data science work, and the baseline for the credit rates above. It is more capable than earlier Sonnet models and draws credits more slowly than Sonnet 5 or Sonnet 4.6. Start with it unless you have a reason to pick something else.
Claude Opus 5.5
A high-capability model for complex architectural decisions, large refactors, and hard debugging, at roughly 2x the credit usage of Sonnet 5.5. Reserve it for tasks that justify the higher consumption.
Older Claude models
Claude Haiku 4.5, Sonnet 5, Sonnet 4.6, Opus 4.6, and Opus 4.8 remain available under More models in the model picker. Each draws credits faster than its 5.5 counterpart. For example, Opus 4.6 and Opus 4.8 both cost more than Opus 5.5. Use them if you have a specific reason to stay on an earlier model.
Choosing a model
A practical pattern:
- Start with Sonnet 5.5 if you are not sure.
- Drop to Haiku 5.5, GLM 5.3 Flash, or DeepSeek V4.1 Flash for shorter, simpler tasks to stretch your credits further.
- Step up to Opus 5.5 when you hit a problem that needs more reasoning, or where a smaller model has already failed.
- Try Kimi K3 or GLM 5.3 for agentic coding, where their speed shows most. Keep in mind that GLM 5.3 cannot see plots, so use a different model for visual data analysis.
For more on how model selection affects your monthly credits, see the FAQ on credit consumption.
Models behind Next Edit Suggestions
Next Edit Suggestions (NES) runs on specialized models hosted by Posit, not on the models above. NES usage is unlimited, is not metered against your credit balance, and is not configurable from Posit Assistant.