Add Qwen3.6-Plus via Fireworks AI (1M context, latest model)

[pollinations-ai]

Model Request: Qwen3.6-Plus (Alibaba)

Provider: Fireworks AI (already available in their API, verified by community)

Release Date: ~April 1-2, 2026 (very recent model)

Key Specs:

• Context Length: 1 million tokens (1M context)
• Output: Up to 65K tokens
• Pricing: Competitive (estimated $0.15-0.30/M input based on Fireworks patterns)
• Strengths: Long-context reasoning, coding, agentic tasks

Use Case:
Requested by marcosfrgames08 for TeleChars AI — a 22-hour roleplay bot that currently struggles with long context windows. The 1M context would enable:

• Full session memory without truncation
• Better character consistency over long interactions
• Reduced token waste on re-summarization

Current Workaround:
Users are forced to use OpenRouter (when payment restrictions allow) or fight with smaller context windows on existing Pollinations models.

Related:

• Issue DeepSeek pricing 2x higher than official API with worse performance via Fireworks #9898 (DeepSeek pricing optimization)
• Issue Add Mistral Small 4 (262K context) as new model, rename current to mistral-fast #9909 (Mistral Small 4 request)

Priority: High — this model fills a gap in Pollinations' lineup (no current model offers 1M context at competitive pricing).

---

Author: marcosfrgames08 (UID: 1117271024271110214)
Source: View on Discord

[wonghk-lab]

Support!

[ElliotEtag]

Thanks for the suggestion — this exact model is already wired up.

• qwen-large in the registry is Qwen 3.6 Plus via Fireworks (accounts/fireworks/models/qwen3p6-plus). Aliases: qwen3.6, qwen3.6-plus.
• 1M context, free for the qualifying tiers.

Closing as resolved.