1stTop pick
Researched 4d agoArrow 2 Telos
- Pick
- Current leader
- Output (from)
- $30.00 / 1M
Best LLM for translation
Last refreshed 2026-09-21. Next refresh: weekly.
Compare multilingual LLMs and dedicated translation models for text, document, and live speech translation. Ranked by context length and general-language benchmarks until translation-specific leaderboard rows land in seed data.
Verdict
Use Arrow 2 Telos for this task today.
GPT-6 Astra is the runner-up: Current leader vs No. 2 on Pick.
2ndShortlistRestricted
Researched 18d agoGPT-6 Astra
- Pick
- No. 2
- Output (from)
- $50.00 / 1M
3rdShortlist
Researched todayXiaomi MiMo-V2.6-Pro
- Pick
- No. 3
- Output (from)
- $0.870 / 1M
How we rank
Translation picks are methodology-forward until dedicated translation benchmark rows land in seed data. We surface tagged translation models, Qwen-MT specialists, and long-context chat models teams commonly route for multilingual work.
- Eligibility — Models with translation use-case tags, translation specialization, Qwen-MT family membership, or ≥128K context on general chat-completion routes (excluding embeddings and media-only specialists).
- Primary ranking — Declared context window (wider first), then MMLU when present, then newer release. Realtime speech translation models appear in the table but are labeled separately from text-generation LLMs.
- Benchmark gap — No translation-quality leaderboard is standardized in seed yet — do not treat this page as a definitive translation-quality ranking until WMT or vendor-neutral multilingual rows are sourced.
- Pricing column — Lowest tracked provider input/output where a public rate card exists.
| # | Model | Input $/1M | Output $/1M | |
|---|---|---|---|---|
| 1 | RWKV-7 Goose 0.1B | — | — | |
| 2 | RWKV-7 Goose 0.4B | — | — | |
| 3 | RWKV-7 Goose 1.5B | — | — | |
| 4 | RWKV-7 Goose 2.9B | — | — | |
| 5 | RWKV-6 Finch 14B | — | — | |
| 6 | RWKV-6 Finch 1.6B | — | — | |
| 7 | RWKV-6 Finch 3B | — | — | |
| 8 | RWKV-6 Finch 7B | — | — | |
| 9 | LTM-2-mini | — | — | |
| 10 | Llama 4 Scout 17B-16E Instruct Vision | $0.08 | $0.22 | |
| 11 | LTM-1 | — | — | |
| 12 | MiniMax-01 Vision | $0.20 | $1.10 | |
| 13 | Gemini 3.5 Pro PreviewReasoningVision | — | — | |
| 14 | Gemini 1.5 Pro 002 | — | — | |
| 15 | Gemini 1.5 Pro Experimental 0827 | — | — | |
| 16 | Gemini 1.5 Pro Experimental 0801 | — | — | |
| 17 | Gemini 1.5 Pro | $1.25 | $5.00 | |
| 18 | GPT-5.5 ReasoningVisionTools | $5.00 | $30.00 | |
| 19 | Arrow 2 Telos VisionTools | $6.00 | $30.00 | |
| 20 | GPT-6 Astra RestrictedReasoningVisionTools | $10.00 | $50.00 |
Honorable mentions
Next seats in this ranking. Lines below are from each model's stored description in LLMReference seed data—spot-check the model page before relying on a capability claim.
Nex-N2.5-Max is Nex AGI's trillion-scale text-only MoE agentic model in the Nex-N2.5 family, released September 8, 2026. Apache-2 weights on Hugging Face (nex-agi/Nex-N2.5-Max). 1.6T total / 49B active MoE with a native 1,048,576-token context window (config max_position_embeddings). Supports reasoning, function calling, and tool use; no vision or multimodal inputs. No OpenRouter listing in this seed pass. Sibling Nex-N2.5-mini is the multimodal MoE variant; Nex-N2.5-Pro is held from this seed pass.
See leaderboard
Rank
Gemini 3.8 Flash is Google DeepMind's generally available most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows, released September 2, 2026. It accepts text, image, video, audio, and PDF inputs and returns text, with a 1,048,576-token context window, up to 65,536 output tokens, and Gemini API support for thinking (low/medium/high; minimal is not supported), function calling, tool use, structured outputs, code execution, prompt caching, search grounding, URL context, computer use (preview), and batch, flex, and priority consumption. Official model ID: gemini-3.8-flash. Sibling of live gemini-3.7-flash under a new family gemini-3.8. Compare it for Coding, Agents, Long context, Vision, and JSON / Tool use.
See leaderboard
Rank
Muse Spark 1.3 is Meta Superintelligence Labs' latest Muse Spark checkpoint, released September 2, 2026. Official Meta Model API model ID: muse-spark-1.3. First-party models docs: the latest version, recommended for new work, tuned for agentic workflows (multi-step tool, browser, and long-horizon tasks) with improved coding over 1.2; default model in docs examples. Product line (dev.meta.ai): Built for long-horizon coding and agentic work. Cleaner code, fewer tokens. First-party research post: rolling out today in Muse Code and Meta Model API; previously available reasoning modes available today with max reasoning coming shortly after additional safety testing. Official table: text, image, video, audio*, PDF input and text output; 1,048,576-token context. Official quickstart limits: context 1048576, output 131072; input modalities text/image/pdf/video, output text. Audio understanding in 1.3 is currently not fully supported (use 1.2 or Muse Voice Transcribe for audio). Tool/function calling, structured outputs, prompt caching, search grounding, and reasoning_effort (minimal/low/medium/high/xhigh; none is not supported). Compare it for Coding, Agents, Long context, and Vision.
See leaderboard
Rank
Compare Top Picks
Side-by-side comparison of the top picks by price, benchmark, and API access.
Browse Other Categories
Best LLMs for Code GenerationBest LLMs for RAGBest LLMs for Agents vs Agent Products (2026)Best LLMs for ClassificationBest Open Source LLMsBest Multimodal / Vision LLMsBest AI Image Models in 2026Best AI Video Models in 2026Best LLMs for Reasoning & MathBest Small Language Models (SLMs)Best LLMs for JSON / Tool useCheapest LLM APIs You Can Call Right NowBest Long Context LLMsBest Mainstream LLM APIs, RankedBest LLMs for EnterpriseBest Free LLMs You Can Use Right NowBest LLMs for WritingBest LLMs for MarketingBest LLMs for Customer Support