Qwen3.8 2.4T A95B vs Devstral 2 123B Instruct 2512
At a Glance
| Compare |  Devstral 2 123B Instruct 2512Mistral AI | |
|---|---|---|
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | $2.00Deepinfra ↗ · Sep 23, 2026 | $0.44Openrouter ↗ · Aug 29, 2026 |
| Output priceFrom · USD / 1M tokens | $6.00Deepinfra ↗ · Sep 23, 2026 | $2.20Openrouter ↗ · Aug 29, 2026 |
| Context windowMaximum documented tokens | 262K | 262K |
| Model facts checked | Aug 28, 2026View model evidence → | Aug 28, 2026View model evidence → |
Token prices are the lowest available sourced USD rates; input and output may use different providers. Cost ranking estimates output spend on LiveBench, not a full request bill. Ranking methodology →
Available Benchmarks
Side-by-Side Facts
| Field | Qwen3.8-2.4T-A95B | Devstral-2-123B-Instruct-2512 |
|---|---|---|
| Developer | Qwen | Mistral AI |
| Family | Qwen3 8 2 4t A95b | Devstral 2 123b Instruct 2512 |
| Model | Qwen3.8-2.4T-A95B | Devstral-2-123B-Instruct-2512 |
| Version | Qwen3.8-2.4T-A95B | Devstral-2-123B-Instruct-2512 |
| Lifecycle | active | active |
| Released | Unknown | Unknown |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
| Context window | 262K | 262K |
| Total parameters | 2.4T | 125B |
| Active parameters | 95B | Unknown |
| License | other | other |
| Open weights | Yes | Yes |
| API available | Yes | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Deepinfra (Standard), Fireworks Ai (Standard), Hugging Face (Standard), Openrouter (Standard), Together Ai (Standard) | Openrouter (Standard) |
| Capabilities | chat, generation, reasoning, tools | chat, generation, tools |
Qwen3.8 2.4T A95B Capabilities
Devstral 2 123B Instruct 2512 Capabilities
Related Comparisons
All text comparisons →| A | Pair | B | Context |
|---|---|---|---|
| vs |  GPT-6 AstraOpenAI | cross-developer peerstext | |
| vs | Claude Fable 5.1Anthropic | cross-developer peerstext | |
| vs |  Gemini 3.1 ProGoogle DeepMind | cross-developer peerstext | |
| vs |  DeepSeek-V4.1-FlashDeepSeek | cross-developer peerstext | |
| vs | Grok 4.6xAI | cross-developer peerstext | |
| vs |  Qwen3.8-MaxQwen | family variantstext | |
| vs |  Kimi-K3Moonshot AI | cross-developer peerstext | |
 MiniMax-M3MiniMax | vs | cross-developer peerstext | |
| vs |  GLM-5.3Z.ai | cross-developer peerstext | |
| vs |  Hy4 previewTencent | cross-developer peerstext | |
| vs | Seed 2.1 ProByteDance Seed | cross-developer peerstext | |
| vs | Mistral Large 3Mistral AI | cross-developer peerstext |
Primary Evidence
Sources and Freshness
Questions
Qwen3.8 2.4T A95B vs Devstral 2 123B Instruct 2512 FAQs
Is Qwen3.8 2.4T A95B or Devstral 2 123B Instruct 2512 better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both Qwen3.8 2.4T A95B and Devstral 2 123B Instruct 2512, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, Qwen3.8 2.4T A95B or Devstral 2 123B Instruct 2512?+
Qwen3.8 2.4T A95B is $2.00 and Devstral 2 123B Instruct 2512 is $0.44 per million tokens, so Devstral 2 123B Instruct 2512 is cheaper on this metric. Qwen3.8 2.4T A95B is $6.00 and Devstral 2 123B Instruct 2512 is $2.20 per million tokens, so Devstral 2 123B Instruct 2512 is cheaper on this metric.
Which has a larger context window, Qwen3.8 2.4T A95B or Devstral 2 123B Instruct 2512?+
Neither model has a larger sourced context window in this comparison. Qwen3.8 2.4T A95B is 262K and Devstral 2 123B Instruct 2512 is 262K.
Which performs better in benchmarks, Qwen3.8 2.4T A95B or Devstral 2 123B Instruct 2512?+
There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.
Can Qwen3.8 2.4T A95B or Devstral 2 123B Instruct 2512 be self-hosted?+
Both models have the same recorded self-hosting status: supported. Qwen3.8 2.4T A95B is open weight; Devstral 2 123B Instruct 2512 is open weight.
Can Qwen3.8 2.4T A95B and Devstral 2 123B Instruct 2512 understand images?+
Qwen3.8 2.4T A95B is not documented with image input; Devstral 2 123B Instruct 2512 is not documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, Qwen3.8 2.4T A95B or Devstral 2 123B Instruct 2512?+
Neither has a larger sourced maximum output. Qwen3.8 2.4T A95B is — and Devstral 2 123B Instruct 2512 is —.
Do Qwen3.8 2.4T A95B and Devstral 2 123B Instruct 2512 support reasoning and tool use?+
Qwen3.8 2.4T A95B: reasoning and tool calling. Devstral 2 123B Instruct 2512: tool calling. Feature support does not establish relative quality.
Which is available from more inference providers, Qwen3.8 2.4T A95B or Devstral 2 123B Instruct 2512?+
Qwen3.8 2.4T A95B has 5 sourced provider routes; Devstral 2 123B Instruct 2512 has 1, so Qwen3.8 2.4T A95B has broader tracked availability.
Which offers better value, Qwen3.8 2.4T A95B or Devstral 2 123B Instruct 2512?+
There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.