Mistral Medium 3.5 128B vs Mistral Large 3
At a Glance
| Compare |  Mistral Medium 3.5 128BMistral AI | Mistral Large 3Mistral AI |
|---|---|---|
| Intelligence, Cost, and Efficiency | ||
| IntelligenceHigher is better · MM Intelligence v4.4 | UnrankedNot in the 57-model eligible cohort | #28 of 5743.0 score · 1/3 sources · single-source provisional · missing LiveBench, ARC-AGI-2 · full-core range 14.3–81.0 |
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | Not reported | $0.25Openrouter ↗ · Oct 19, 2026 |
| Output priceFrom · USD / 1M tokens | Not reported | $0.75Openrouter ↗ · Oct 19, 2026 |
| Context windowMaximum documented tokens | 262K | 262K |
| Model facts checked | Aug 28, 2026View model evidence → | Aug 29, 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
| Benchmark | Mistral-Medium-3.5-128B | Mistral Large 3 |
|---|---|---|
| LMArena Text Arenatext-2026-10-15-0ebb411abf00 · arena_rating · statistical tie | 1,420.9799% of row best · rating · mistral-medium-3.5; 95% CI [1414.59009121, 1427.34230551]; votes 11778; rank 112 | 1,428.16100% of row best · rating · mistral-large-3; 95% CI [1425.33671505, 1430.99053174]; votes 84205; rank 98 |
| LMArena Vision Arenavision-2026-10-09-0ebb411abf00 · arena_rating · statistical tie | 1,223.08100% of row best · rating · mistral-medium-3.5; 95% CI [1213.57754699, 1232.58266173]; votes 5375; rank 82 | 1,220.76100% of row best · rating · mistral-large-3; 95% CI [1212.81813492, 1228.69335345]; votes 8643; rank 83 |
| Overall ResultCounted from the protocol-matched rows above · 2 ties | 0 benchmark winsNo overall winner | 0 benchmark winsNo overall winner |
Third-party benchmark Only like-for-like primary-publisher results are shown; raw scores, relative scores, configuration, and token spend remain visible.
Side-by-Side Facts
| Field | Mistral-Medium-3.5-128B | Mistral Large 3 |
|---|---|---|
| Developer | Mistral AI | Mistral AI |
| Family | Mistral Medium 3 5 128b | Mistral Large 3 |
| Model | Mistral-Medium-3.5-128B | Mistral Large 3 |
| Version | Mistral-Medium-3.5-128B | Mistral Large 3 |
| Lifecycle | active | active |
| Released | Unknown | 2025-12-02 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text, Image | Text, Image, Document |
| Output modalities | Text | Text |
| Context window | 262K | 262K |
| Total parameters | 127.7B | 675B |
| Active parameters | Unknown | 41B |
| License | other | Apache-2.0 |
| Open weights | Yes | Yes |
| API available | Unknown | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Unknown | Mistral AI (Standard), Openrouter (Standard) |
| Capabilities | chat, generation, reasoning, tools | agents, chat, generation, structured_outputs, tools, vision |
Mistral Medium 3.5 128B Capabilities
Mistral Large 3 Capabilities
Related Comparisons
All image comparisons →| A | Pair | B | Context |
|---|---|---|---|
| vs | Mistral-Medium-3.5-128BMistral AI | cross-developer peersimage, text | |
| vs | Mistral-Medium-3.5-128BMistral AI | cross-developer peersimage, text | |
| vs | Mistral-Medium-3.5-128BMistral AI | cross-developer peersimage, text | |
Mistral-Medium-3.5-128BMistral AI | vs | Mistral-Small-4-119B-2603Mistral AI | family variantsimage, text |
Mistral-Medium-3.5-128BMistral AI | vs |  Kimi-K2.6Moonshot AI | cross-developer peersimage, text |
Mistral-Medium-3.5-128BMistral AI | vs | cross-developer peerstext | |
 Qwen3-Coder-NextQwen | vs | Mistral-Medium-3.5-128BMistral AI | cross-developer peerstext |
Qwen3.5-35B-A3BQwen | vs | Mistral-Medium-3.5-128BMistral AI | cross-developer peersimage, text |
| vs | Mistral-Medium-3.5-128BMistral AI | cross-developer peersimage, text |
Primary Evidence
Sources and Freshness
Questions
Mistral Medium 3.5 128B vs Mistral Large 3 FAQs
Is Mistral Medium 3.5 128B or Mistral Large 3 better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both Mistral Medium 3.5 128B and Mistral Large 3, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, Mistral Medium 3.5 128B or Mistral Large 3?+
Only Mistral Large 3 has a directly sourced input price: $0.25 per million tokens. Only Mistral Large 3 has a directly sourced output price: $0.75 per million tokens.
Which has a larger context window, Mistral Medium 3.5 128B or Mistral Large 3?+
Both models have the same sourced context window: 262,144 tokens each.
Which performs better in benchmarks, Mistral Medium 3.5 128B or Mistral Large 3?+
There is no overall benchmark winner: Neither model leads outside the score tolerance on the verified common benchmarks.
Can Mistral Medium 3.5 128B or Mistral Large 3 be self-hosted?+
Both models have the same sourced self-hosting status: supported. Mistral Medium 3.5 128B is open weight; Mistral Large 3 is open weight.
Can Mistral Medium 3.5 128B and Mistral Large 3 understand images?+
Mistral Medium 3.5 128B is documented with image input; Mistral Large 3 is documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, Mistral Medium 3.5 128B or Mistral Large 3?+
Neither model has a sourced maximum output in this comparison, so a larger limit cannot be determined.
Do Mistral Medium 3.5 128B and Mistral Large 3 support reasoning and tool use?+
Mistral Medium 3.5 128B: reasoning, tool calling, and image input. Mistral Large 3: tool calling and image input. Feature support does not establish relative quality.
Which is available from more inference providers, Mistral Medium 3.5 128B or Mistral Large 3?+
Mistral Medium 3.5 128B has 0 sourced provider routes; Mistral Large 3 has 2, so Mistral Large 3 has broader tracked availability.
Which offers better value, Mistral Medium 3.5 128B or Mistral Large 3?+
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.